------------------------------------------------------------------------------------------------------------------------
      name:  <unnamed>
       log:  /Users/marziaoceno/Dropbox/How Social Desirability Bias Impacts the Expression of Emotions.log
  log type:  text
 opened on:  10 May 2025, 14:38:15

. 
. ** Author: Marzia Oceno

. ** Title: "How Social Desirability Bias Impacts the Expression of Emotions"

. ** Journal: Political Science Research and Methods

. ** This file recreates the following from the paper:

. 
. *******************************ANES 2012*****************************************

. *********************

. ** Figure 1 -- 2012 **

. *********************

. 
. cibar anger [aweight=weight_full], over(web partyid_2cat) graphopts(title("Anger") ytitle("Mean Anger toward Out-Party
>  Cand.") scheme(plotplain) name(anger12, replace))

. cibar fear [aweight=weight_full], over(web partyid_2cat) graphopts(title("Fear") ytitle("Mean Fear toward Out-Party Ca
> nd.")  scheme(plotplain) name(fear12, replace))

. cibar hopeful [aweight=weight_full], over(web partyid_2cat) graphopts(title("Hope") ytitle("Mean Hope toward In-Party 
> Candid.")  scheme(plotplain) name(hope12, replace))

. cibar proud [aweight=weight_full], over(web partyid_2cat) graphopts(title("Pride") ytitle("Mean Pride toward In-Party 
> Candid.")  scheme(plotplain) name(pride12, replace))

. 
. graph combine anger12 fear12 hope12 pride12, scheme(plotplain) xcommon ycommon col(2)

. 
. 
. *********************

. ** Figure 3 -- 2012 **

. *********************

. 
. svyset [pweight=weight_full], strata(strata_full) psu(psu_full)

Sampling weights: weight_full
             VCE: linearized
     Single unit: missing
        Strata 1: strata_full
 Sampling unit 1: psu_full
           FPC 1: <zero>

. 
. svy: reg anger web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,759
Number of PSUs   = 1,905                           Population size = 2,455.861
                                                   Design df       =     1,895
                                                   F(10, 1886)     =     14.22
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.0871

-------------------------------------------------------------------------------
              |             Linearized
        anger | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0604935   .0203879     2.97   0.003     .0205085    .1004786
  campaignint |   .2184908   .0340288     6.42   0.000      .151753    .2852287
       polint |   .1284386   .0445211     2.88   0.004     .0411231     .215754
      polknow |  -.0143426   .0513695    -0.28   0.780    -.1150893    .0864042
       gender |    .041748   .0191062     2.19   0.029     .0042767    .0792194
        White |   .0097661   .0231202     0.42   0.673    -.0355776    .0551098
       latinx |  -.0218243   .0309666    -0.70   0.481    -.0825564    .0389079
         ageN |  -.1934399   .0565036    -3.42   0.001    -.3042558   -.0826241
    education |   .0474676   .0370762     1.28   0.201    -.0252469    .1201822
income_norm01 |  -.0241114   .0387397    -0.62   0.534    -.1000884    .0518656
        _cons |   .1861288   .0476253     3.91   0.000     .0927254    .2795322
-------------------------------------------------------------------------------

. estimates store angryDe

. svy: reg anger web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,836
Number of PSUs   = 1,489                           Population size = 2,121.787
                                                   Design df       =     1,479
                                                   F(10, 1470)     =     25.74
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1352

-------------------------------------------------------------------------------
              |             Linearized
        anger | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0807825   .0193493     4.17   0.000     .0428275    .1187374
  campaignint |   .1687951   .0367129     4.60   0.000     .0967802      .24081
       polint |    .208203   .0496239     4.20   0.000     .1108623    .3055437
      polknow |  -.0149873   .0522918    -0.29   0.774    -.1175612    .0875867
       gender |  -.0050757   .0175449    -0.29   0.772    -.0394912    .0293399
        White |   .0948241   .0431216     2.20   0.028     .0102381      .17941
       latinx |   .0335752   .0548915     0.61   0.541    -.0740983    .1412487
         ageN |   .1017905   .0604643     1.68   0.092    -.0168144    .2203954
    education |  -.0659144   .0368009    -1.79   0.073    -.1381019    .0062731
income_norm01 |    .020241   .0356664     0.57   0.570     -.049721    .0902031
        _cons |    .066775   .0563613     1.18   0.236    -.0437816    .1773316
-------------------------------------------------------------------------------

. estimates store angryRe

. coefplot (angryDe) || (angryRe), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN education inco
> me_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Anger toward Out-Party Candidate") mlabe
> l format(%9.3f) mlabsize(medsmall) mlabposition(12) name(angerplot12)

. 
. svy: reg fear web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,762
Number of PSUs   = 1,905                           Population size = 2,459.058
                                                   Design df       =     1,895
                                                   F(10, 1886)     =     11.78
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.0664

-------------------------------------------------------------------------------
              |             Linearized
         fear | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |    .052467    .019553     2.68   0.007     .0141194    .0908147
  campaignint |   .1708656   .0311969     5.48   0.000     .1096818    .2320495
       polint |   .1249002   .0451076     2.77   0.006     .0364344    .2133661
      polknow |  -.0681471   .0478613    -1.42   0.155    -.1620134    .0257192
       gender |    .015721   .0185945     0.85   0.398    -.0207468    .0521888
        White |   .0139536   .0223959     0.62   0.533    -.0299696    .0578768
       latinx |   -.067328   .0273262    -2.46   0.014    -.1209207   -.0137353
         ageN |  -.0367766   .0558098    -0.66   0.510    -.1462317    .0726785
    education |   .0280592   .0347867     0.81   0.420    -.0401651    .0962835
income_norm01 |    .015249   .0388116     0.39   0.694     -.060869     .091367
        _cons |   .1091621   .0447696     2.44   0.015     .0213593     .196965
-------------------------------------------------------------------------------

. estimates store afraidDe

. svy: reg fear web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,838
Number of PSUs   = 1,490                           Population size = 2,124.153
                                                   Design df       =     1,480
                                                   F(10, 1471)     =     25.30
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1237

-------------------------------------------------------------------------------
              |             Linearized
         fear | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .1236523   .0199834     6.19   0.000     .0844536     .162851
  campaignint |   .1603995   .0379182     4.23   0.000     .0860204    .2347785
       polint |   .1569721   .0494941     3.17   0.002      .059886    .2540583
      polknow |    .039603    .053042     0.75   0.455    -.0644425    .1436485
       gender |   .0343117   .0194745     1.76   0.078    -.0038888    .0725121
        White |   .0785247   .0422424     1.86   0.063    -.0043366    .1613859
       latinx |   .0170786   .0562263     0.30   0.761    -.0932131    .1273704
         ageN |   .1544932   .0604455     2.56   0.011     .0359252    .2730612
    education |  -.0618392   .0403049    -1.53   0.125    -.1409001    .0172216
income_norm01 |   .0095667   .0381423     0.25   0.802     -.065252    .0843854
        _cons |  -.0857731   .0543123    -1.58   0.114    -.1923104    .0207641
-------------------------------------------------------------------------------

. estimates store afraidRe

. coefplot (afraidDe) || (afraidRe), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN education in
> come_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Fear toward Out-Party Candidate") mlab
> el format(%9.3f) mlabsize(medsmall) mlabposition(12) name(fearplot12)

. 
. svy: reg hopeful web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,768
Number of PSUs   = 1,903                           Population size = 2,463.194
                                                   Design df       =     1,893
                                                   F(10, 1884)     =     21.73
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1397

-------------------------------------------------------------------------------
              |             Linearized
      hopeful | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |  -.0564577   .0182821    -3.09   0.002    -.0923129   -.0206024
  campaignint |   .1655221   .0273455     6.05   0.000     .1118916    .2191527
       polint |   .1624914   .0353207     4.60   0.000     .0932198     .231763
      polknow |   .0114417   .0401681     0.28   0.776    -.0673368    .0902201
       gender |   .0423603   .0157956     2.68   0.007     .0113818    .0733389
        White |  -.1379781   .0190662    -7.24   0.000     -.175371   -.1005852
       latinx |  -.0642323   .0255115    -2.52   0.012    -.1142659   -.0141987
         ageN |  -.0167868   .0458604    -0.37   0.714     -.106729    .0731554
    education |   .0119494   .0312784     0.38   0.702    -.0493943    .0732932
income_norm01 |   .0065117   .0318737     0.20   0.838    -.0559996    .0690231
        _cons |   .4177186   .0432538     9.66   0.000     .3328883    .5025488
-------------------------------------------------------------------------------

. estimates store hopefulDe

. svy: reg hopeful web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,832
Number of PSUs   = 1,489                           Population size = 2,117.583
                                                   Design df       =     1,479
                                                   F(10, 1470)     =     22.66
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1580

-------------------------------------------------------------------------------
              |             Linearized
      hopeful | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0097969   .0210202     0.47   0.641    -.0314357    .0510294
  campaignint |   .2169213    .036768     5.90   0.000     .1447982    .2890443
       polint |   .1796436   .0485619     3.70   0.000     .0843861    .2749011
      polknow |   .0146573   .0491513     0.30   0.766    -.0817564    .1110709
       gender |   .0635293   .0175517     3.62   0.000     .0291004    .0979582
        White |   .0558678   .0475416     1.18   0.240    -.0373883    .1491239
       latinx |   .0761422   .0615633     1.24   0.216    -.0446184    .1969028
         ageN |   .1429005   .0613606     2.33   0.020     .0225374    .2632637
    education |  -.0763444   .0382885    -1.99   0.046      -.15145   -.0012388
income_norm01 |   .0720697   .0350025     2.06   0.040       .00341    .1407294
        _cons |   .0324221    .069825     0.46   0.642    -.1045445    .1693887
-------------------------------------------------------------------------------

. estimates store hopefulRe

. coefplot (hopefulDe) || (hopefulRe), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN education 
> income_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Hope toward In-Party Candidate") mla
> bel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(hopeplot12)

. 
. svy: reg proud web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,763
Number of PSUs   = 1,903                           Population size = 2,459.566
                                                   Design df       =     1,893
                                                   F(10, 1884)     =     38.38
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1940

-------------------------------------------------------------------------------
              |             Linearized
        proud | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |  -.0549685   .0197912    -2.78   0.006    -.0937834   -.0161536
  campaignint |   .2072205   .0298975     6.93   0.000      .148585    .2658561
       polint |   .1934217   .0377299     5.13   0.000     .1194251    .2674182
      polknow |  -.0195274   .0415734    -0.47   0.639     -.101062    .0620072
       gender |    .045237    .016654     2.72   0.007     .0125749     .077899
        White |  -.2100214   .0201465   -10.42   0.000    -.2495332   -.1705097
       latinx |  -.1556803   .0267129    -5.83   0.000    -.2080701   -.1032905
         ageN |   .0519563   .0504416     1.03   0.303    -.0469707    .1508832
    education |  -.0036547   .0313537    -0.12   0.907    -.0651462    .0578368
income_norm01 |   .0286753   .0322089     0.89   0.373    -.0344934    .0918439
        _cons |   .4024261   .0434749     9.26   0.000     .3171622    .4876899
-------------------------------------------------------------------------------

. estimates store proudDe

. svy: reg proud web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,823
Number of PSUs   = 1,490                           Population size = 2,101.327
                                                   Design df       =     1,480
                                                   F(10, 1471)     =     32.77
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1760

-------------------------------------------------------------------------------
              |             Linearized
        proud | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0474718    .023509     2.02   0.044     .0013573    .0935863
  campaignint |   .2345256   .0394658     5.94   0.000     .1571108    .3119404
       polint |   .2043627   .0511975     3.99   0.000     .1039352    .3047901
      polknow |  -.0252561   .0549819    -0.46   0.646    -.1331068    .0825947
       gender |   .1120389   .0191052     5.86   0.000     .0745628    .1495151
        White |   .0689706   .0419219     1.65   0.100    -.0132621    .1512033
       latinx |   .0742721   .0562438     1.32   0.187    -.0360539    .1845981
         ageN |   .1718569   .0627263     2.74   0.006      .048815    .2948988
    education |  -.0409834   .0382022    -1.07   0.284    -.1159197    .0339529
income_norm01 |   .0068304   .0349014     0.20   0.845     -.061631    .0752918
        _cons |  -.1502946   .0557804    -2.69   0.007    -.2597116   -.0408775
-------------------------------------------------------------------------------

. estimates store proudRe

. coefplot (proudDe) || (proudRe), scheme(plottig) legend(off) drop(?cons ageN campaignint polint polknow education inco
> me_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Pride toward In-Party Candidate") mlabel
>  format(%9.3f) mlabsize(medsmall) mlabposition(12) name(prideplot12)

. 
. graph combine angerplot12 fearplot12 hopeplot12 prideplot12, scheme(plottig) xcommon ycommon col(2)

. 
. 
. *********************

. ** Figure 5 -- 2012 **

. *********************

. 
. svy: logit persuade c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,288.471
                                                   Design df       =     1,770
                                                   F(12, 1759)     =     15.08
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.000948   .2849785     3.51   0.000     .4420183    1.559878
        1.web |   .1778652    .206506     0.86   0.389    -.2271561    .5828864
              |
  web#c.anger |
           1  |  -.3669051   .3463839    -1.06   0.290     -1.04627    .3124595
              |
  campaignint |    1.48007   .2369267     6.25   0.000     1.015385    1.944756
       polint |   1.332827   .2869763     4.64   0.000     .7699786    1.895675
      polknow |  -.2137451   .3153266    -0.68   0.498    -.8321968    .4047065
       gender |  -.0704108   .1257511    -0.56   0.576    -.3170472    .1762255
        White |  -.2852514   .1478975    -1.93   0.054    -.5753235    .0048208
       latinx |  -.3351057   .2022054    -1.66   0.098    -.7316922    .0614809
         ageN |   .4554457   .3961999     1.15   0.250    -.3216232    1.232515
    education |  -.3754531   .2369555    -1.58   0.113    -.8401952     .089289
income_norm01 |   .4719472   .2366366     1.99   0.046     .0078307    .9360638
        _cons |  -2.553599   .3252943    -7.85   0.000      -3.1916   -1.915598
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file1D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,288.471
Model VCE: Linearized                              Design df       =     1,770

Expression: Pr(persuade), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .1962209   .0525661     3.73   0.000     .0931227    .2993191
          2  |   .1277491   .0409336     3.12   0.002     .0474659    .2080323
------------------------------------------------------------------------------

. svy: logit rally c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,572
Number of PSUs   = 1,780                           Population size = 2,288.169
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      7.20
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .4898116    .447532     1.09   0.274    -.3879353    1.367558
        1.web |   .0323276   .4199736     0.08   0.939    -.7913688     .856024
              |
  web#c.anger |
           1  |  -.0184238   .5952964    -0.03   0.975    -1.185982    1.149134
              |
  campaignint |   2.006626   .4513077     4.45   0.000     1.121474    2.891778
       polint |   .7411165   .5712955     1.30   0.195    -.3793683    1.861601
      polknow |   .0241461   .5705061     0.04   0.966    -1.094791    1.143083
       gender |  -.0725974    .249202    -0.29   0.771    -.5613585    .4161637
        White |   -.679304   .2096912    -3.24   0.001    -1.090572   -.2680356
       latinx |  -1.176752   .4520148    -2.60   0.009    -2.063291   -.2902133
         ageN |   .4346705   .7976541     0.54   0.586    -1.129773    1.999114
    education |    1.29451   .4444615     2.91   0.004     .4227853    2.166235
income_norm01 |  -.2473522   .4198194    -0.59   0.556    -1.070746    .5760417
        _cons |  -5.194451   .9375082    -5.54   0.000    -7.033191   -3.355711
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file2D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,572
Number of PSUs   = 1,780                           Population size = 2,288.169
Model VCE: Linearized                              Design df       =     1,770

Expression: Pr(rally), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |    .027374   .0249036     1.10   0.272    -.0214695    .0762175
          2  |    .026827   .0204309     1.31   0.189    -.0132443    .0668982
------------------------------------------------------------------------------

. svy: logit button c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyi
> d_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,288.471
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      8.62
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .9910783   .3223973     3.07   0.002     .3587589    1.623398
        1.web |   .4348459   .2697735     1.61   0.107    -.0942623     .963954
              |
  web#c.anger |
           1  |  -.1853941   .4175661    -0.44   0.657    -1.004369    .6335803
              |
  campaignint |   1.532672   .3106589     4.93   0.000      .923375    2.141969
       polint |   .4409262   .3463474     1.27   0.203    -.2383667    1.120219
      polknow |   -.113183   .3629758    -0.31   0.755    -.8250894    .5987233
       gender |  -.0260055   .1618597    -0.16   0.872    -.3434618    .2914508
        White |  -.9367899   .1630407    -5.75   0.000    -1.256563   -.6170173
       latinx |  -.7571445   .2305721    -3.28   0.001    -1.209367   -.3049222
         ageN |   .6723597   .4670242     1.44   0.150    -.2436173    1.588337
    education |  -.4837513   .2769574    -1.75   0.081    -1.026949    .0594467
income_norm01 |   .2966654   .2728786     1.09   0.277    -.2385328    .8318635
        _cons |  -3.091563   .4709028    -6.57   0.000    -4.015147   -2.167979
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file3D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,288.471
Model VCE: Linearized                              Design df       =     1,770

Expression: Pr(button), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |      .1145   .0355337     3.22   0.001     .0448076    .1841924
          2  |   .1131562   .0357977     3.16   0.002      .042946    .1833663
------------------------------------------------------------------------------

. svy: logit volunteer c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if par
> tyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,288.471
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      5.74
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.245106   .5748576     2.17   0.030     .1176351    2.372577
        1.web |    .186315   .4724389     0.39   0.693    -.7402818    1.112912
              |
  web#c.anger |
           1  |  -.3081413   .7248005    -0.43   0.671    -1.729696    1.113414
              |
  campaignint |   1.421751   .5743389     2.48   0.013     .2952972    2.548205
       polint |   1.266233   .7157865     1.77   0.077    -.1376428    2.670109
      polknow |   .0214423   .6667188     0.03   0.974    -1.286197    1.329081
       gender |  -.1348868    .290096    -0.46   0.642    -.7038535    .4340799
        White |  -.5912431   .2787647    -2.12   0.034    -1.137986   -.0445003
       latinx |  -.5203619   .4801858    -1.08   0.279    -1.462153    .4214289
         ageN |   1.931107   .9674569     2.00   0.046     .0336287    3.828585
    education |   .7552936    .469118     1.61   0.108    -.1647899    1.675377
income_norm01 |   .2527119   .4362437     0.58   0.562    -.6028952    1.108319
        _cons |  -6.633148   1.189095    -5.58   0.000    -8.965326   -4.300969
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file4D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,288.471
Model VCE: Linearized                              Design df       =     1,770

Expression: Pr(volunteer), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .0509491   .0260508     1.96   0.051    -.0001445    .1020427
          2  |   .0390529   .0171448     2.28   0.023     .0054267     .072679
------------------------------------------------------------------------------

. svy: logit donation c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,571
Number of PSUs   = 1,780                           Population size = 2,287.711
                                                   Design df       =     1,770
                                                   F(12, 1759)     =     10.93
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .8101405   .4380256     1.85   0.065    -.0489614    1.669242
        1.web |     .08887   .3324687     0.27   0.789    -.5632026    .7409426
              |
  web#c.anger |
           1  |   .4634777   .5410634     0.86   0.392    -.5977126    1.524668
              |
  campaignint |   1.468052   .3490984     4.21   0.000     .7833639    2.152741
       polint |   1.022937   .4248781     2.41   0.016     .1896219    1.856253
      polknow |   .5309804   .4984081     1.07   0.287    -.4465499    1.508511
       gender |  -.0974029    .189597    -0.51   0.608    -.4692605    .2744548
        White |  -.7850236    .201773    -3.89   0.000    -1.180762   -.3892852
       latinx |  -.5126506   .3108013    -1.65   0.099    -1.122227    .0969256
         ageN |   3.286107   .6225569     5.28   0.000     2.065083    4.507131
    education |    1.04503   .3324316     3.14   0.002     .3930306     1.69703
income_norm01 |   1.597129   .3556278     4.49   0.000     .8996348    2.294624
        _cons |  -7.090438   .7736289    -9.17   0.000     -8.60776   -5.573116
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file5D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,571
Number of PSUs   = 1,780                           Population size = 2,287.711
Model VCE: Linearized                              Design df       =     1,770

Expression: Pr(donation), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .0675403   .0355237     1.90   0.057    -.0021325    .1372132
          2  |   .1231268   .0308157     4.00   0.000     .0626879    .1835657
------------------------------------------------------------------------------

. 
. combomarginsplot "file5D" "file4D" "file3D" "file2D" "file1D", labels("Donate" "Volunteer" "Display sign" "Attend rall
> y" "Persuade others") recast(scatter) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("M
> arginal Effect of Anger by Mode among Dems") title("") legend(pos(6) row(1)) xsize(8) ysize(6) xlabel(-.6(.2).6) plot1
> opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) 
> ci2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(anger12Dem)

  Variables that uniquely identify margins: web _filenumber

. 
. svy: logit persuade c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
                                                   Design df       =     1,378
                                                   F(12, 1367)     =     15.67
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.104662   .4782444     2.31   0.021     .1664959    2.042828
        1.web |  -.4209978   .2843176    -1.48   0.139      -.97874    .1367444
              |
  web#c.anger |
           1  |   .5445071   .5384748     1.01   0.312    -.5118119    1.600826
              |
  campaignint |   1.240042    .279547     4.44   0.000      .691658    1.788425
       polint |   .9333729   .3751283     2.49   0.013     .1974886    1.669257
      polknow |   .7812633   .3704113     2.11   0.035     .0546323    1.507894
       gender |  -.0101028   .1413929    -0.07   0.943    -.2874713    .2672658
        White |    .593357   .2566475     2.31   0.021     .0898948    1.096819
       latinx |   .8204393   .3763994     2.18   0.029     .0820614    1.558817
         ageN |   .0981278   .4816874     0.20   0.839    -.8467921    1.043048
    education |   -.007393   .2726387    -0.03   0.978    -.5422247    .5274388
income_norm01 |   .3783991    .279617     1.35   0.176     -.170122    .9269202
        _cons |  -3.306137   .4628174    -7.14   0.000     -4.21404   -2.398234
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file1R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
Model VCE: Linearized                              Design df       =     1,378

Expression: Pr(persuade), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |    .232612   .0934399     2.49   0.013     .0493122    .4159118
          2  |   .3331236   .0469377     7.10   0.000     .2410465    .4252006
------------------------------------------------------------------------------

. svy: logit rally c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      6.60
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .5689928   .9131699     0.62   0.533    -1.222361    2.360346
        1.web |  -.0785164   .5993188    -0.13   0.896    -1.254192    1.097159
              |
  web#c.anger |
           1  |   .1527747   1.036642     0.15   0.883    -1.880792    2.186342
              |
  campaignint |   .9413843   .8643409     1.09   0.276    -.7541819    2.636951
       polint |   1.905424   .9158462     2.08   0.038     .1088209    3.702028
      polknow |   .8960942   .9155447     0.98   0.328    -.8999179    2.692106
       gender |   .3545847   .2574662     1.38   0.169    -.1504833    .8596527
        White |  -.3251093   .4214698    -0.77   0.441    -1.151901    .5016826
       latinx |   .1615584   .7202463     0.22   0.823    -1.251339    1.574456
         ageN |     .39337   1.128675     0.35   0.728    -1.820736    2.607476
    education |   .8729293   .4701263     1.86   0.064    -.0493114     1.79517
income_norm01 |  -.7438118   .5119954    -1.45   0.147    -1.748187     .260563
        _cons |  -6.077841   1.007764    -6.03   0.000    -8.054757   -4.100924
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file2R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
Model VCE: Linearized                              Design df       =     1,378

Expression: Pr(rally), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .0280402   .0472772     0.59   0.553    -.0647028    .1207832
          2  |   .0357974   .0267988     1.34   0.182    -.0167735    .0883683
------------------------------------------------------------------------------

. svy: logit button c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyi
> d_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      8.67
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.509399   .6007244     2.51   0.012     .3309656    2.687832
        1.web |   .5184399   .4253077     1.22   0.223    -.3158808    1.352761
              |
  web#c.anger |
           1  |  -.8941779   .6785599    -1.32   0.188      -2.2253    .4369443
              |
  campaignint |   1.332825   .4251186     3.14   0.002     .4988755    2.166775
       polint |   1.840732   .5139273     3.58   0.000     .8325677    2.848897
      polknow |  -.2807667   .5488841    -0.51   0.609    -1.357505     .795972
       gender |   .2442395   .1835598     1.33   0.184    -.1158474    .6043264
        White |  -.0865676   .3238022    -0.27   0.789    -.7217661    .5486309
       latinx |   -.085823    .524415    -0.16   0.870    -1.114561    .9429151
         ageN |  -.2226635   .6372925    -0.35   0.727    -1.472832    1.027505
    education |  -.4345283    .374269    -1.16   0.246    -1.168727    .2996702
income_norm01 |  -.6541931   .3509891    -1.86   0.063    -1.342724    .0343376
        _cons |  -4.113432   .6675685    -6.16   0.000    -5.422992   -2.803871
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file3R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
Model VCE: Linearized                              Design df       =     1,378

Expression: Pr(button), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .1623061   .0684866     2.37   0.018     .0279567    .2966554
          2  |   .0682035   .0401149     1.70   0.089    -.0104893    .1468963
------------------------------------------------------------------------------

. svy: logit volunteer c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if par
> tyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      5.72
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.023475   1.293739     0.79   0.429    -1.514437    3.561386
        1.web |   .4688513   1.005084     0.47   0.641    -1.502809    2.440511
              |
  web#c.anger |
           1  |  -.5296643   1.529271    -0.35   0.729    -3.529615    2.470287
              |
  campaignint |   .9176879   .8071331     1.14   0.256    -.6656545     2.50103
       polint |   2.677664   .9308742     2.88   0.004     .8515801    4.503748
      polknow |  -.0703519   1.291189    -0.05   0.957    -2.603261    2.462557
       gender |  -.0259289   .4095092    -0.06   0.950    -.8292579       .7774
        White |  -.1144153   .5606749    -0.20   0.838    -1.214284    .9854534
       latinx |  -.5231716   .7509117    -0.70   0.486    -1.996225    .9498821
         ageN |  -.0990947   1.769158    -0.06   0.955    -3.569628    3.371439
    education |   .2793235    .540336     0.52   0.605    -.7806466    1.339294
income_norm01 |  -2.123718   .6238833    -3.40   0.001    -3.347582   -.8998545
        _cons |   -5.85565    1.43698    -4.07   0.000    -8.674555   -3.036745
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file4R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
Model VCE: Linearized                              Design df       =     1,378

Expression: Pr(volunteer), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |    .023061   .0317346     0.73   0.468    -.0391923    .0853143
          2  |   .0126998   .0205921     0.62   0.538    -.0276953     .053095
------------------------------------------------------------------------------

. svy: logit donation c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      9.04
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.051457   .7737883     1.36   0.174    -.4664735    2.569387
        1.web |   .1820186   .5010553     0.36   0.716    -.8008951    1.164932
              |
  web#c.anger |
           1  |  -.1692108   .8410452    -0.20   0.841    -1.819078    1.480657
              |
  campaignint |   1.245248   .5561912     2.24   0.025      .154175    2.336321
       polint |   2.419008    .554988     4.36   0.000     1.330295    3.507721
      polknow |   1.812027   .6799838     2.66   0.008     .4781116    3.145942
       gender |   .2872639   .2125904     1.35   0.177    -.1297718    .7042996
        White |  -.5432975   .3622597    -1.50   0.134    -1.253938    .1673426
       latinx |  -.4064512   .5523007    -0.74   0.462    -1.489892    .6769899
         ageN |   1.243527   .7923763     1.57   0.117    -.3108669    2.797922
    education |    .486355   .3730979     1.30   0.193    -.2455462    1.218256
income_norm01 |   .6554249   .4092824     1.60   0.110    -.1474591    1.458309
        _cons |  -7.746195   .9233856    -8.39   0.000    -9.557588   -5.934801
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file5R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
Model VCE: Linearized                              Design df       =     1,378

Expression: Pr(donation), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .0877743    .066469     1.32   0.187    -.0426171    .2181657
          2  |   .0776711   .0317837     2.44   0.015     .0153214    .1400208
------------------------------------------------------------------------------

. 
. combomarginsplot "file5R" "file4R" "file3R" "file2R" "file1R", labels("Donate" "Volunteer" "Display sign" "Attend rall
> y" "Persuade others") recast(scatter) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("M
> arginal Effect of Anger by Mode among Reps") title("") legend(pos(6) row(1)) xsize(8) ysize(6) xlabel(-.6(.2).6) plot1
> opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) 
> ci2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(anger12Rep)

  Variables that uniquely identify margins: web _filenumber

. 
. 
. svy: logit persuade c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,574
Number of PSUs   = 1,780                           Population size = 2,286.622
                                                   Design df       =     1,770
                                                   F(12, 1759)     =     15.32
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.150689   .2921312     3.94   0.000     .5777304    1.723647
        1.web |   .1786673    .183769     0.97   0.331    -.1817598    .5390945
              |
   web#c.fear |
           1  |   -.467355    .351627    -1.33   0.184    -1.157003    .2222929
              |
  campaignint |    1.51767   .2335696     6.50   0.000     1.059568    1.975771
       polint |   1.343466   .2846857     4.72   0.000     .7851106    1.901822
      polknow |   -.193451     .31221    -0.62   0.536      -.80579    .4188881
       gender |  -.0561052   .1246556    -0.45   0.653    -.3005929    .1883825
        White |  -.2954651   .1484088    -1.99   0.047    -.5865401   -.0043901
       latinx |  -.3160159   .2040973    -1.55   0.122    -.7163131    .0842813
         ageN |   .3455646   .3904831     0.88   0.376     -.420292    1.111421
    education |  -.3688208   .2363175    -1.56   0.119    -.8323115    .0946699
income_norm01 |   .4436726   .2356187     1.88   0.060    -.0184476    .9057927
        _cons |  -2.504448    .305797    -8.19   0.000    -3.104209   -1.904686
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file6D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,574
Number of PSUs   = 1,780                           Population size = 2,286.622
Model VCE: Linearized                              Design df       =     1,770

Expression: Pr(persuade), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .2237579    .052916     4.23   0.000     .1199734    .3275424
          2  |   .1374472   .0391855     3.51   0.000     .0605924     .214302
------------------------------------------------------------------------------

. svy: logit rally c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_
> 2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,574
Number of PSUs   = 1,780                           Population size = 2,291.233
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      5.80
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .6510495   .4827514     1.35   0.178    -.2957734    1.597872
        1.web |  -.0034542   .3343612    -0.01   0.992    -.6592386    .6523301
              |
   web#c.fear |
           1  |  -.2300797   .5875166    -0.39   0.695    -1.382379    .9222195
              |
  campaignint |   1.481003   .6285218     2.36   0.019     .2482801    2.713726
       polint |   .9601058    .602521     1.59   0.111    -.2216217    2.141833
      polknow |   .0383597   .5609811     0.07   0.945    -1.061895    1.138615
       gender |  -.1426372    .249584    -0.57   0.568    -.6321476    .3468732
        White |  -.6153872   .2146969    -2.87   0.004    -1.036473   -.1943011
       latinx |  -1.228348   .4532981    -2.71   0.007    -2.117404   -.3392922
         ageN |  -.0554976   .8534717    -0.07   0.948    -1.729416    1.618421
    education |   .9973306   .5018848     1.99   0.047     .0129814     1.98168
income_norm01 |   .0089145   .4704631     0.02   0.985    -.9138072    .9316363
        _cons |  -4.533858   .9814305    -4.62   0.000    -6.458742   -2.608973
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file7D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,574
Number of PSUs   = 1,780                           Population size = 2,291.233
Model VCE: Linearized                              Design df       =     1,770

Expression: Pr(rally), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .0402242   .0325932     1.23   0.217    -.0237011    .1041495
          2  |   .0242438   .0190527     1.27   0.203    -.0131243     .061612
------------------------------------------------------------------------------

. svy: logit button c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,575
Number of PSUs   = 1,780                           Population size = 2,291.535
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      9.20
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .8718486   .3377918     2.58   0.010     .2093358    1.534361
        1.web |   .2379301   .2382932     1.00   0.318    -.2294355    .7052957
              |
   web#c.fear |
           1  |   .0884316   .4231199     0.21   0.834    -.7414357    .9182989
              |
  campaignint |   1.398541   .3408679     4.10   0.000     .7299954    2.067087
       polint |   .5506537   .3523762     1.56   0.118    -.1404635    1.241771
      polknow |  -.0856225   .3623448    -0.24   0.813    -.7962912    .6250462
       gender |  -.0417525   .1616437    -0.26   0.796    -.3587852    .2752803
        White |  -.9347788   .1661845    -5.62   0.000    -1.260717   -.6088402
       latinx |  -.7433261    .229689    -3.24   0.001    -1.193816   -.2928359
         ageN |    .392118   .4938092     0.79   0.427    -.5763926    1.360629
    education |  -.5830938   .2958492    -1.97   0.049    -1.163344   -.0028433
income_norm01 |   .3518379   .2823273     1.25   0.213    -.2018921    .9055678
        _cons |  -2.709578   .4761507    -5.69   0.000    -3.643455   -1.775701
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file8D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,575
Number of PSUs   = 1,780                           Population size = 2,291.535
Model VCE: Linearized                              Design df       =     1,770

Expression: Pr(button), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .1059809   .0400923     2.64   0.008     .0273477     .184614
          2  |   .1344393    .034804     3.86   0.000      .066178    .2027006
------------------------------------------------------------------------------

. svy: logit volunteer c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,575
Number of PSUs   = 1,780                           Population size = 2,291.535
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      5.65
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .8160819   .5925434     1.38   0.169    -.3460765     1.97824
        1.web |   .1198022    .405616     0.30   0.768    -.6757345     .915339
              |
   web#c.fear |
           1  |  -.2597675   .7265523    -0.36   0.721    -1.684758    1.165223
              |
  campaignint |   1.570046   .5853029     2.68   0.007      .422088    2.718003
       polint |   1.364655   .7148476     1.91   0.056     -.037379     2.76669
      polknow |  -.0331229   .6709975    -0.05   0.961    -1.349154    1.282908
       gender |  -.1109767   .2873475    -0.39   0.699    -.6745528    .4525994
        White |   -.598726    .275805    -2.17   0.030    -1.139664   -.0577882
       latinx |  -.5485716   .4732485    -1.16   0.247    -1.476756    .3796132
         ageN |   1.822006   .9458556     1.93   0.054    -.0331052    3.677118
    education |   .7512863   .4636122     1.62   0.105    -.1579986    1.660571
income_norm01 |   .2369363   .4299802     0.55   0.582     -.606386    1.080259
        _cons |  -6.421128   1.172456    -5.48   0.000    -8.720672   -4.121585
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file9D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,575
Number of PSUs   = 1,780                           Population size = 2,291.535
Model VCE: Linearized                              Design df       =     1,770

Expression: Pr(volunteer), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |    .033719   .0255815     1.32   0.188    -.0164542    .0838921
          2  |   .0232235   .0156689     1.48   0.138     -.007508    .0539549
------------------------------------------------------------------------------

. svy: logit donation c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,290.775
                                                   Design df       =     1,770
                                                   F(12, 1759)     =     12.67
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.231789   .4140905     2.97   0.003      .419631    2.043947
        1.web |   .3589385   .3037741     1.18   0.238    -.2368552    .9547321
              |
   web#c.fear |
           1  |  -.1372147   .5102512    -0.27   0.788    -1.137973    .8635437
              |
  campaignint |   1.502305    .354017     4.24   0.000     .8079691     2.19664
       polint |   1.072888    .436224     2.46   0.014     .2173198    1.928456
      polknow |   .5779862   .5133003     1.13   0.260    -.4287524    1.584725
       gender |  -.1000945   .1899204    -0.53   0.598    -.4725863    .2723973
        White |  -.7753275   .2034941    -3.81   0.000    -1.174442   -.3762134
       latinx |  -.4832293      .3093    -1.56   0.118    -1.089861    .1234024
         ageN |    3.12489   .6174924     5.06   0.000     1.913799    4.335981
    education |   1.027634   .3319533     3.10   0.002     .3765724    1.678696
income_norm01 |   1.603158   .3453902     4.64   0.000     .9257423    2.280573
        _cons |  -7.195305   .7256746    -9.92   0.000    -8.618574   -5.772035
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file10D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,290.775
Model VCE: Linearized                              Design df       =     1,770

Expression: Pr(donation), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .1014206   .0323499     3.14   0.002     .0379725    .1648687
          2  |   .1060712   .0298441     3.55   0.000     .0475377    .1646046
------------------------------------------------------------------------------

. 
. combomarginsplot "file10D" "file9D" "file8D" "file7D" "file6D", labels("Donate" "Volunteer" "Display sign" "Attend ral
> ly" "Persuade others") recast(scatter) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("
> Marginal Effect of Fear by Mode among Dems") title("") legend(pos(6) row(1)) xsize(8) ysize(6) xlabel(-.6(.2).6) plot1
> opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) 
> ci2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(fear12Dem)

  Variables that uniquely identify margins: web _filenumber

. 
. svy: logit persuade c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
                                                   Design df       =     1,379
                                                   F(12, 1368)     =     14.43
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .6737774   .4515579     1.49   0.136    -.2120374    1.559592
        1.web |  -.4411361   .2346545    -1.88   0.060    -.9014546    .0191824
              |
   web#c.fear |
           1  |   .7811883   .5058279     1.54   0.123     -.211087    1.773464
              |
  campaignint |   1.308765   .2811618     4.65   0.000     .7572136    1.860316
       polint |    1.00852   .3661111     2.75   0.006     .2903248    1.726715
      polknow |   .6396292    .377596     1.69   0.091    -.1010956    1.380354
       gender |  -.0525927   .1402368    -0.38   0.708    -.3276933    .2225078
        White |   .6171972    .253829     2.43   0.015     .1192645     1.11513
       latinx |   .8490518   .3921001     2.17   0.031     .0798746    1.618229
         ageN |   .0405783   .4914826     0.08   0.934    -.9235562    1.004713
    education |  -.0304296    .270976    -0.11   0.911    -.5619993    .5011401
income_norm01 |   .3889073   .2734765     1.42   0.155    -.1475677    .9253823
        _cons |  -3.021267    .429024    -7.04   0.000    -3.862877   -2.179656
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file6R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
Model VCE: Linearized                              Design df       =     1,379

Expression: Pr(persuade), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .1445318   .0936184     1.54   0.123    -.0391181    .3281818
          2  |   .2949986   .0435403     6.78   0.000     .2095863    .3804109
------------------------------------------------------------------------------

. svy: logit rally c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_
> 2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      6.47
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.293185   .7225721     1.79   0.074    -.1242749    2.710644
        1.web |   .5259651   .4919089     1.07   0.285    -.4390057    1.490936
              |
   web#c.fear |
           1  |  -1.178744   .8580556    -1.37   0.170     -2.86198    .5044911
              |
  campaignint |   1.021426   .8262964     1.24   0.217    -.5995082     2.64236
       polint |   1.952373   .9051437     2.16   0.031     .1767652     3.72798
      polknow |    .950974   .9016614     1.05   0.292    -.8178023     2.71975
       gender |   .3440257   .2704806     1.27   0.204    -.1865723    .8746237
        White |  -.2821064   .4254635    -0.66   0.507    -1.116732    .5525193
       latinx |   .1912371   .7204026     0.27   0.791    -1.221966    1.604441
         ageN |   .3610532   1.130449     0.32   0.749    -1.856533     2.57864
    education |   .8710918   .4872718     1.79   0.074    -.0847823    1.826966
income_norm01 |  -.7484524   .5133034    -1.46   0.145    -1.755392    .2584875
        _cons |  -6.472773   .9455745    -6.85   0.000    -8.327693   -4.617853
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file7R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
Model VCE: Linearized                              Design df       =     1,379

Expression: Pr(rally), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .0670849   .0433541     1.55   0.122    -.0179621     .152132
          2  |   .0057796   .0271444     0.21   0.831    -.0474693    .0590284
------------------------------------------------------------------------------

. svy: logit button c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      8.64
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.364539    .447409     3.05   0.002     .4868629    2.242215
        1.web |   .3529496   .3152251     1.12   0.263     -.265423    .9713222
              |
   web#c.fear |
           1  |  -.8326974   .5336994    -1.56   0.119    -1.879648    .2142531
              |
  campaignint |   1.391341   .4039565     3.44   0.001     .5989055    2.183777
       polint |    1.85852   .5055731     3.68   0.000     .8667446    2.850296
      polknow |  -.3403807    .541649    -0.63   0.530    -1.402926    .7221643
       gender |   .1928221   .1840572     1.05   0.295    -.1682403    .5538845
        White |  -.0760733   .3152496    -0.24   0.809     -.694494    .5423475
       latinx |  -.0805887   .5287278    -0.15   0.879    -1.117787    .9566092
         ageN |  -.2373279    .630532    -0.38   0.707    -1.474234    .9995778
    education |   -.439941   .3726177    -1.18   0.238      -1.1709    .2910178
income_norm01 |  -.6724265   .3525821    -1.91   0.057    -1.364082    .0192287
        _cons |  -3.838898   .6288496    -6.10   0.000    -5.072503   -2.605292
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file8R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
Model VCE: Linearized                              Design df       =     1,379

Expression: Pr(button), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .1496413       .053     2.82   0.005      .045672    .2536106
          2  |   .0587175   .0343889     1.71   0.088    -.0087426    .1261776
------------------------------------------------------------------------------

. svy: logit volunteer c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      5.14
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.253707   1.492212     0.84   0.401    -1.673544    4.180958
        1.web |   .0166701   .9663474     0.02   0.986       -1.879     1.91234
              |
   web#c.fear |
           1  |   .0412075   1.589338     0.03   0.979    -3.076575     3.15899
              |
  campaignint |   .8088518   .7858562     1.03   0.304    -.7327512    2.350455
       polint |   2.659002   .9821166     2.71   0.007     .7323977    4.585606
      polknow |  -.2505869   1.325876    -0.19   0.850    -2.851538    2.350364
       gender |  -.1066345   .4244404    -0.25   0.802    -.9392532    .7259841
        White |  -.0787657   .5943932    -0.13   0.895    -1.244778    1.087247
       latinx |  -.4565339   .7875244    -0.58   0.562    -2.001409    1.088342
         ageN |   -.256133   1.787047    -0.14   0.886    -3.761757    3.249491
    education |   .3634774   .5436772     0.67   0.504    -.7030464    1.430001
income_norm01 |  -2.223438   .6362215    -3.49   0.000    -3.471505   -.9753714
        _cons |  -5.494324   1.496663    -3.67   0.000    -8.430307   -2.558342
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file9R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
Model VCE: Linearized                              Design df       =     1,379

Expression: Pr(volunteer), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .0297887   .0379877     0.78   0.433    -.0447312    .1043087
          2  |   .0318387   .0163349     1.95   0.051    -.0002052    .0638825
------------------------------------------------------------------------------

. svy: logit donation c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      9.11
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .5077391   .5869131     0.87   0.387    -.6436001    1.659078
        1.web |  -.0317586   .3559731    -0.09   0.929    -.7300659    .6665488
              |
   web#c.fear |
           1  |   .2470781   .6483762     0.38   0.703    -1.024832    1.518988
              |
  campaignint |   1.335338   .5396922     2.47   0.013     .2766316    2.394044
       polint |    2.46245   .5547592     4.44   0.000     1.374186    3.550713
      polknow |   1.738384   .6810018     2.55   0.011     .4024724    3.074295
       gender |   .2428156    .212441     1.14   0.253    -.1739268     .659558
        White |  -.5560727   .3607539    -1.54   0.123    -1.263759    .1516131
       latinx |  -.4216921   .5384011    -0.78   0.434    -1.477866    .6344816
         ageN |   1.256605   .7850037     1.60   0.110    -.2833253    2.796536
    education |   .4726119   .3656721     1.29   0.196    -.2447218    1.189946
income_norm01 |    .667887   .4136091     1.61   0.107    -.1434841    1.479258
        _cons |  -7.426187   .8062679    -9.21   0.000    -9.007831   -5.844543
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file10R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
Model VCE: Linearized                              Design df       =     1,379

Expression: Pr(donation), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .0424684   .0501578     0.85   0.397    -.0559255    .1408623
          2  |   .0660579   .0273825     2.41   0.016     .0123421    .1197736
------------------------------------------------------------------------------

. 
. combomarginsplot "file10R" "file9R" "file8R" "file7R" "file6R", labels("Donate" "Volunteer" "Display sign" "Attend ral
> ly" "Persuade others") recast(scatter) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("
> Marginal Effect of Fear by Mode among Reps") title("") legend(pos(6) row(1)) xsize(8) ysize(6) xlabel(-.6(.2).6) plot1
> opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) 
> ci2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(fear12Rep)

  Variables that uniquely identify margins: web _filenumber

. 
. 
. svy: logit persuade c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if pa
> rtyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,580
Number of PSUs   = 1,778                           Population size = 2,290.758
                                                   Design df       =     1,768
                                                   F(12, 1757)     =     13.68
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   .1030272   .3848212     0.27   0.789    -.6517251    .8577796
        1.web |  -.0108515   .2862426    -0.04   0.970     -.572261     .550558
              |
web#c.hopeful |
           1  |   .1856627   .4458197     0.42   0.677    -.6887264    1.060052
              |
  campaignint |    1.59667   .2402241     6.65   0.000     1.125517    2.067823
       polint |   1.401529    .286492     4.89   0.000       .83963    1.963427
      polknow |  -.2499356   .3170882    -0.79   0.431    -.8718427    .3719716
       gender |  -.0484427   .1250961    -0.39   0.699    -.2937945    .1969092
        White |  -.2519815   .1507255    -1.67   0.095    -.5476004    .0436373
       latinx |  -.3487115   .2003353    -1.74   0.082    -.7416305    .0442075
         ageN |   .3544806   .3917601     0.90   0.366    -.4138811    1.122842
    education |  -.3212425   .2371419    -1.35   0.176    -.7863504    .1438654
income_norm01 |   .4319908   .2398324     1.80   0.072    -.0383942    .9023757
        _cons |  -2.343624   .3628109    -6.46   0.000    -3.055208   -1.632041
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file11D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,580
Number of PSUs   = 1,778                           Population size = 2,290.758
Model VCE: Linearized                              Design df       =     1,768

Expression: Pr(persuade), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0207209   .0772136     0.27   0.788    -.1307186    .1721604
          2  |   .0585561   .0522255     1.12   0.262    -.0438741    .1609862
------------------------------------------------------------------------------

. svy: logit rally c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,580
Number of PSUs   = 1,778                           Population size = 2,295.369
                                                   Design df       =     1,768
                                                   F(12, 1757)     =      5.40
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |    .886592   .9125824     0.97   0.331    -.9032619    2.676446
        1.web |   .1027892   .7843816     0.13   0.896    -1.435624    1.641202
              |
web#c.hopeful |
           1  |   -.207837   1.037355    -0.20   0.841    -2.242408    1.826734
              |
  campaignint |   1.445783   .6340778     2.28   0.023     .2021621    2.689404
       polint |   .9357577   .5748141     1.63   0.104     -.191629    2.063144
      polknow |   .0427671   .5592423     0.08   0.939    -1.054079    1.139613
       gender |   -.155069   .2469242    -0.63   0.530     -.639363     .329225
        White |  -.5171923   .2417324    -2.14   0.033    -.9913036   -.0430811
       latinx |  -1.197889   .4510212    -2.66   0.008     -2.08248   -.3132984
         ageN |  -.0819729   .8496745    -0.10   0.923    -1.748445    1.584499
    education |   1.037908   .5068216     2.05   0.041     .0438751     2.03194
income_norm01 |   -.001155   .4775441    -0.00   0.998    -.9377655    .9354556
        _cons |  -4.885737   1.087115    -4.49   0.000    -7.017902   -2.753572
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file12D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,580
Number of PSUs   = 1,778                           Population size = 2,295.369
Model VCE: Linearized                              Design df       =     1,768

Expression: Pr(rally), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0526665   .0517663     1.02   0.309    -.0488632    .1541961
          2  |   .0395882   .0328367     1.21   0.228    -.0248146     .103991
------------------------------------------------------------------------------

. svy: logit button c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,581
Number of PSUs   = 1,778                           Population size = 2,295.671
                                                   Design df       =     1,768
                                                   F(12, 1757)     =      8.81
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.151578   .4996229     2.30   0.021     .1716644    2.131492
        1.web |   .3366906   .4200057     0.80   0.423    -.4870694    1.160451
              |
web#c.hopeful |
           1  |   .0735758   .5948812     0.12   0.902    -1.093169     1.24032
              |
  campaignint |   1.376699    .337428     4.08   0.000     .7148992    2.038499
       polint |   .5052783   .3393219     1.49   0.137    -.1602359    1.170793
      polknow |  -.1470205    .358804    -0.41   0.682    -.8507452    .5567041
       gender |  -.0776539   .1612867    -0.48   0.630    -.3939866    .2386788
        White |  -.7516835    .169415    -4.44   0.000    -1.083958   -.4194087
       latinx |  -.7289724   .2293294    -3.18   0.002    -1.178758   -.2791871
         ageN |    .365506   .4908528     0.74   0.457    -.5972069    1.328219
    education |  -.5527573   .2968303    -1.86   0.063    -1.134933     .029418
income_norm01 |   .3736191   .2848038     1.31   0.190    -.1849686    .9322067
        _cons |  -3.160929   .5482638    -5.77   0.000    -4.236243   -2.085616
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file13D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,581
Number of PSUs   = 1,778                           Population size = 2,295.671
Model VCE: Linearized                              Design df       =     1,768

Expression: Pr(button), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .1342813   .0563497     2.38   0.017     .0237623    .2448003
          2  |   .1751068   .0494386     3.54   0.000     .0781426     .272071
------------------------------------------------------------------------------

. svy: logit volunteer c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if p
> artyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,581
Number of PSUs   = 1,778                           Population size = 2,295.671
                                                   Design df       =     1,768
                                                   F(12, 1757)     =      5.90
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   2.641767   .8736331     3.02   0.003     .9283048     4.35523
        1.web |   1.378612   .8081909     1.71   0.088    -.2064983    2.963722
              |
web#c.hopeful |
           1  |  -1.827246   1.073705    -1.70   0.089    -3.933111    .2786188
              |
  campaignint |   1.499913   .5529928     2.71   0.007     .4153249    2.584502
       polint |   1.279412   .7112452     1.80   0.072    -.1155581    2.674382
      polknow |  -.0411177   .6740414    -0.06   0.951     -1.36312    1.280884
       gender |  -.1434584   .2873695    -0.50   0.618    -.7070781    .4201613
        White |  -.3973204   .3098164    -1.28   0.200    -1.004965    .2103246
       latinx |  -.4503591   .4803372    -0.94   0.349    -1.392448    .4917295
         ageN |   1.787044   .9257218     1.93   0.054    -.0285805    3.602668
    education |   .7882865   .4846656     1.63   0.104    -.1622913    1.738864
income_norm01 |   .2679158   .4454599     0.60   0.548    -.6057676    1.141599
        _cons |  -7.940318   1.376127    -5.77   0.000    -10.63932   -5.241311
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file14D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,581
Number of PSUs   = 1,778                           Population size = 2,295.671
Model VCE: Linearized                              Design df       =     1,768

Expression: Pr(volunteer), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0992463   .0352448     2.82   0.005     .0301204    .1683721
          2  |   .0346055   .0297942     1.16   0.246      -.02383    .0930411
------------------------------------------------------------------------------

. svy: logit donation c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if pa
> rtyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,579
Number of PSUs   = 1,778                           Population size = 2,294.911
                                                   Design df       =     1,768
                                                   F(12, 1757)     =     11.62
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   2.032446    .730778     2.78   0.005     .5991657    3.465725
        1.web |   .5653841   .6154953     0.92   0.358     -.641791    1.772559
              |
web#c.hopeful |
           1  |  -.2223099   .8346169    -0.27   0.790     -1.85925     1.41463
              |
  campaignint |   1.436407   .3643125     3.94   0.000     .7218781    2.150935
       polint |   .9542465   .4321063     2.21   0.027     .1067535    1.801739
      polknow |   .5772959   .5068819     1.14   0.255    -.4168549    1.571447
       gender |  -.1268133   .1900951    -0.67   0.505    -.4996481    .2460214
        White |  -.5589616   .2068943    -2.70   0.007    -.9647448   -.1531784
       latinx |  -.4710042   .3067484    -1.54   0.125    -1.072632    .1306236
         ageN |   3.048115    .611295     4.99   0.000     1.849178    4.247052
    education |   1.124728   .3473225     3.24   0.001     .4435218    1.805934
income_norm01 |   1.625833   .3518288     4.62   0.000     .9357889    2.315877
        _cons |  -8.057492   .8510527    -9.47   0.000    -9.726667   -6.388316
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file15D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,579
Number of PSUs   = 1,778                           Population size = 2,294.911
Model VCE: Linearized                              Design df       =     1,768

Expression: Pr(donation), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .1594156   .0533987     2.99   0.003     .0546843    .2641469
          2  |   .1773274   .0450235     3.94   0.000     .0890225    .2656324
------------------------------------------------------------------------------

. 
. combomarginsplot "file15D" "file14D" "file13D" "file12D" "file11D", labels("Donate" "Volunteer" "Display sign" "Attend
>  rally" "Persuade others") recast(scatter) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtit
> le("Marginal Effect of Hope by Mode among Dems") title("") legend(pos(6) row(1)) xsize(8) ysize(6) xlabel(-.6(.2).6) p
> lot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(
> D)) ci2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(hope12Dem)

  Variables that uniquely identify margins: web _filenumber

. 
. svy: logit persuade c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if pa
> rtyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
                                                   Design df       =     1,378
                                                   F(12, 1367)     =     14.79
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.077136   .3990042     2.70   0.007     .2944142    1.859857
        1.web |   -.189227   .2779329    -0.68   0.496    -.7344444    .3559904
              |
web#c.hopeful |
           1  |   .3246405   .4730419     0.69   0.493    -.6033195    1.252601
              |
  campaignint |   1.267798   .2815654     4.50   0.000     .7154546    1.820141
       polint |   .9040886   .3580323     2.53   0.012     .2017412    1.606436
      polknow |   .6463024   .3708085     1.74   0.082    -.0811079    1.373713
       gender |  -.1062252   .1395978    -0.76   0.447    -.3800723     .167622
        White |    .590599   .2555122     2.31   0.021     .0893639    1.091834
       latinx |   .6827616   .3707513     1.84   0.066    -.0445364     1.41006
         ageN |   .0152418   .4758337     0.03   0.974    -.9181951    .9486786
    education |   -.010632   .2679348    -0.04   0.968    -.5362362    .5149721
income_norm01 |   .3000846   .2720119     1.10   0.270    -.2335177    .8336869
        _cons |  -3.132272   .4509849    -6.95   0.000    -4.016964   -2.247581
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file11R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
Model VCE: Linearized                              Design df       =     1,378

Expression: Pr(persuade), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .2275841   .0784852     2.90   0.004     .0736208    .3815474
          2  |   .2891068   .0515378     5.61   0.000     .1880058    .3902078
------------------------------------------------------------------------------

. svy: logit rally c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      6.92
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.452342   1.032148     1.41   0.160    -.5724093    3.477093
        1.web |   .8722277    .745673     1.17   0.242    -.5905494    2.335005
              |
web#c.hopeful |
           1  |  -1.366687   1.123163    -1.22   0.224    -3.569982    .8366074
              |
  campaignint |   .9816364   .8123048     1.21   0.227    -.6118513    2.575124
       polint |   1.974247   .8962066     2.20   0.028     .2161704    3.732324
      polknow |   .9511691   .9015172     1.06   0.292    -.8173254    2.719664
       gender |   .3332092   .2704702     1.23   0.218    -.1973688    .8637871
        White |  -.2903336   .4286546    -0.68   0.498     -1.13122    .5505526
       latinx |   .1749789   .7047025     0.25   0.804    -1.207427    1.557385
         ageN |    .349152   1.097933     0.32   0.751    -1.804648    2.502952
    education |   .8697571     .47066     1.85   0.065    -.0535305    1.793045
income_norm01 |   -.747213   .4997449    -1.50   0.135    -1.727556    .2331301
        _cons |   -6.77285   .9945013    -6.81   0.000    -8.723751    -4.82195
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file12R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
Model VCE: Linearized                              Design df       =     1,378

Expression: Pr(rally), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0687396   .0499196     1.38   0.169    -.0291871    .1666662
          2  |    .004357   .0271781     0.16   0.873    -.0489579     .057672
------------------------------------------------------------------------------

. svy: logit button c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      7.02
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   2.090116   .6826642     3.06   0.002     .7509429     3.42929
        1.web |   1.183822    .508496     2.33   0.020     .1863123    2.181332
              |
web#c.hopeful |
           1  |  -1.827522    .780256    -2.34   0.019     -3.35814   -.2969041
              |
  campaignint |   1.369632   .4071979     3.36   0.001     .5708369    2.168426
       polint |   1.868025   .4972684     3.76   0.000       .89254     2.84351
      polknow |  -.2886761   .5522453    -0.52   0.601    -1.372009    .7946562
       gender |   .1761754   .1829921     0.96   0.336    -.1827978    .5351486
        White |  -.0550565   .3143555    -0.18   0.861    -.6717237    .5616107
       latinx |  -.1182606   .5364665    -0.22   0.826     -1.17064    .9341188
         ageN |  -.3084147   .6297833    -0.49   0.624    -1.543852     .927023
    education |   -.452007   .3669832    -1.23   0.218    -1.171913    .2678991
income_norm01 |  -.6691551   .3393868    -1.97   0.049    -1.334926   -.0033845
        _cons |  -4.542773   .8142368    -5.58   0.000    -6.140051   -2.945495
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file13R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
Model VCE: Linearized                              Design df       =     1,378

Expression: Pr(button), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .2086925   .0686465     3.04   0.002     .0740295    .3433555
          2  |   .0297348   .0390256     0.76   0.446    -.0468212    .1062908
------------------------------------------------------------------------------

. svy: logit volunteer c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if p
> artyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      4.69
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   2.164182   .8660481     2.50   0.013      .465267    3.863098
        1.web |   1.426062   .7712893     1.85   0.065    -.0869667     2.93909
              |
web#c.hopeful |
           1  |  -1.926334   1.086311    -1.77   0.076    -4.057336    .2046677
              |
  campaignint |   .9298247   .7378042     1.26   0.208    -.5175162    2.377166
       polint |    2.70171    .944538     2.86   0.004     .8488216    4.554598
      polknow |  -.0420047   1.270959    -0.03   0.974    -2.535229    2.451219
       gender |  -.0677544    .406433    -0.17   0.868    -.8650488      .72954
        White |   -.076756   .5472967    -0.14   0.888    -1.150381    .9968688
       latinx |  -.5984203    .742327    -0.81   0.420    -2.054633     .857793
         ageN |  -.2093345   1.748829    -0.12   0.905     -3.63999    3.221321
    education |   .2876302   .5406202     0.53   0.595    -.7728973    1.348158
income_norm01 |  -2.144826   .6047874    -3.55   0.000    -3.331229   -.9584219
        _cons |    -6.6496   1.368853    -4.86   0.000    -9.334861   -3.964339
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file14R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
Model VCE: Linearized                              Design df       =     1,378

Expression: Pr(volunteer), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0454466   .0252235     1.80   0.072    -.0040339    .0949271
          2  |   .0062067   .0193381     0.32   0.748    -.0317285     .044142
------------------------------------------------------------------------------

. svy: logit donation c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if pa
> rtyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      8.70
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.369575   .6160342     2.22   0.026     .1611085    2.578041
        1.web |   .3831643   .5115728     0.75   0.454    -.6203814     1.38671
              |
web#c.hopeful |
           1  |  -.3493733   .7478229    -0.47   0.640    -1.816368    1.117621
              |
  campaignint |    1.28202   .5279637     2.43   0.015     .2463202    2.317719
       polint |    2.39997   .5529774     4.34   0.000     1.315201    3.484738
      polknow |   1.800588   .6681534     2.69   0.007     .4898805    3.111296
       gender |   .2180344   .2071797     1.05   0.293    -.1883873     .624456
        White |  -.5784066   .3708864    -1.56   0.119     -1.30597    .1491565
       latinx |  -.5084925   .5194629    -0.98   0.328    -1.527516    .5105312
         ageN |   1.082382   .7720348     1.40   0.161     -.432109    2.596872
    education |   .5041413   .3661907     1.38   0.169    -.2142102    1.222493
income_norm01 |   .6098255   .4103913     1.49   0.138    -.1952337    1.414885
        _cons |  -7.840065   .8835575    -8.87   0.000    -9.573329   -6.106802
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file15R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
Model VCE: Linearized                              Design df       =     1,378

Expression: Pr(donation), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |    .109928   .0493213     2.23   0.026     .0131751    .2066809
          2  |   .0914448   .0362174     2.52   0.012     .0203976     .162492
------------------------------------------------------------------------------

. 
. combomarginsplot "file15R" "file14R" "file13R" "file12R" "file11R", labels("Donate" "Volunteer" "Display sign" "Attend
>  rally" "Persuade others") recast(scatter) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtit
> le("Marginal Effect of Hope by Mode among Reps") title("") legend(pos(6) row(1)) xsize(8) ysize(6) xlabel(-.6(.2).6) p
> lot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(
> D)) ci2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(hope12Rep)

  Variables that uniquely identify margins: web _filenumber

. 
. 
. svy: logit persuade c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,576
Number of PSUs   = 1,778                           Population size = 2,287.619
                                                   Design df       =     1,768
                                                   F(12, 1757)     =     13.79
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .3107015   .3275326     0.95   0.343    -.3316903    .9530933
        1.web |  -.0192214   .2742144    -0.07   0.944    -.5570399    .5185971
              |
  web#c.proud |
           1  |   .2026434   .3829945     0.53   0.597    -.5485262    .9538131
              |
  campaignint |   1.544469   .2388548     6.47   0.000     1.076001    2.012936
       polint |   1.326281   .2878043     4.61   0.000     .7618087    1.890754
      polknow |  -.2436974    .317259    -0.77   0.443    -.8659396    .3785448
       gender |  -.0682009   .1242182    -0.55   0.583     -.311831    .1754291
        White |  -.1847576   .1518453    -1.22   0.224    -.4825728    .1130575
       latinx |   -.300267   .2032055    -1.48   0.140    -.6988152    .0982812
         ageN |   .3280736   .3918337     0.84   0.403    -.4404325     1.09658
    education |  -.3157471   .2374105    -1.33   0.184    -.7813819    .1498877
income_norm01 |   .4056887   .2390172     1.70   0.090    -.0630974    .8744748
        _cons |  -2.387569   .3574824    -6.68   0.000    -3.088701   -1.686436
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file16D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,576
Number of PSUs   = 1,778                           Population size = 2,287.619
Model VCE: Linearized                              Design df       =     1,768

Expression: Pr(persuade), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .0626633   .0650269     0.96   0.335    -.0648744     .190201
          2  |    .104041   .0452027     2.30   0.021     .0153846    .1926974
------------------------------------------------------------------------------

. svy: logit rally c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                            Number of obs   =    2,576
Number of PSUs   = 1,778                            Population size = 2,292.23
                                                    Design df       =    1,768
                                                    F(12, 1757)     =     5.35
                                                    Prob > F        =   0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |  -.1375704   .5967503    -0.23   0.818    -1.307981     1.03284
        1.web |    -.62519   .5523065    -1.13   0.258    -1.708432    .4580525
              |
  web#c.proud |
           1  |   .9185847   .7015039     1.31   0.191    -.4572795    2.294449
              |
  campaignint |   1.502269   .6276178     2.39   0.017     .2713177     2.73322
       polint |   .9371956   .5800637     1.62   0.106    -.2004872    2.074878
      polknow |   .0273626   .5547066     0.05   0.961    -1.060587    1.115312
       gender |  -.1537464   .2490327    -0.62   0.537     -.642176    .3346832
        White |  -.5304242   .2399574    -2.21   0.027    -1.001054   -.0597942
       latinx |  -1.206655   .4564816    -2.64   0.008    -2.101955   -.3113547
         ageN |  -.0805989   .8449311    -0.10   0.924    -1.737768     1.57657
    education |   1.068506   .4989507     2.14   0.032     .0899111    2.047102
income_norm01 |  -.0467349   .4754537    -0.10   0.922    -.9792453    .8857756
        _cons |  -4.258855   1.008515    -4.22   0.000    -6.236862   -2.280847
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file17D")

Average marginal effects

Number of strata =    10                            Number of obs   =    2,576
Number of PSUs   = 1,778                            Population size = 2,292.23
Model VCE: Linearized                               Design df       =    1,768

Expression: Pr(rally), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |  -.0084963    .037247    -0.23   0.820    -.0815492    .0645565
          2  |   .0455054   .0253275     1.80   0.073    -.0041695    .0951803
------------------------------------------------------------------------------

. svy: logit button c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyi
> d_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,577
Number of PSUs   = 1,778                           Population size = 2,292.532
                                                   Design df       =     1,768
                                                   F(12, 1757)     =      9.78
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.457926   .5233162     2.79   0.005     .4315427     2.48431
        1.web |   .3795068    .451833     0.84   0.401    -.5066762     1.26569
              |
  web#c.proud |
           1  |   .0539436   .5855393     0.09   0.927    -1.094478    1.202366
              |
  campaignint |   1.274746   .3412959     3.74   0.000     .6053597    1.944131
       polint |   .4603494   .3528711     1.30   0.192    -.2317392    1.152438
      polknow |  -.1019925   .3578308    -0.29   0.776    -.8038085    .5998235
       gender |  -.0760572   .1619117    -0.47   0.639    -.3936158    .2415014
        White |  -.6188097   .1681529    -3.68   0.000     -.948609   -.2890103
       latinx |   -.608421   .2314781    -2.63   0.009    -1.062421   -.1544213
         ageN |   .2592877   .5010351     0.52   0.605    -.7233958    1.241971
    education |  -.5316174   .2980363    -1.78   0.075    -1.116158    .0529232
income_norm01 |   .3326818   .2875484     1.16   0.247    -.2312887    .8966523
        _cons |   -3.34278   .5904098    -5.66   0.000    -4.500755   -2.184806
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file18D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,577
Number of PSUs   = 1,778                           Population size = 2,292.532
Model VCE: Linearized                              Design df       =     1,768

Expression: Pr(button), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .1663129   .0547162     3.04   0.002     .0589976    .2736282
          2  |   .2141937   .0409539     5.23   0.000     .1338705    .2945168
------------------------------------------------------------------------------

. svy: logit volunteer c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if par
> tyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,577
Number of PSUs   = 1,778                           Population size = 2,292.532
                                                   Design df       =     1,768
                                                   F(12, 1757)     =      5.29
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |      2.423   .9238438     2.62   0.009     .6110594    4.234941
        1.web |   1.027549   .8090373     1.27   0.204    -.5592208     2.61432
              |
  web#c.proud |
           1  |  -1.243482   1.016667    -1.22   0.221    -3.237478    .7505145
              |
  campaignint |   1.388416   .5481597     2.53   0.011     .3133064    2.463525
       polint |   1.212651   .7092158     1.71   0.087    -.1783392     2.60364
      polknow |   .0223124   .6766205     0.03   0.974    -1.304748    1.349373
       gender |  -.1615218   .2912547    -0.55   0.579    -.7327615     .409718
        White |  -.3449849   .2960685    -1.17   0.244     -.925666    .2356961
       latinx |  -.3828592   .4778551    -0.80   0.423     -1.32008    .5543611
         ageN |   1.740943   .9401213     1.85   0.064    -.1029237    3.584809
    education |   .8431962   .4846309     1.74   0.082    -.1073137    1.793706
income_norm01 |   .2077191   .4451562     0.47   0.641    -.6653687    1.080807
        _cons |  -7.743419   1.417488    -5.46   0.000    -10.52355    -4.96329
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file19D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,577
Number of PSUs   = 1,778                           Population size = 2,292.532
Model VCE: Linearized                              Design df       =     1,768

Expression: Pr(volunteer), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .0913045   .0361419     2.53   0.012     .0204191    .1621898
          2  |   .0502881   .0216056     2.33   0.020      .007913    .0926633
------------------------------------------------------------------------------

. svy: logit donation c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,575
Number of PSUs   = 1,778                           Population size = 2,291.772
                                                   Design df       =     1,768
                                                   F(12, 1757)     =     12.54
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .6873078   .5699299     1.21   0.228    -.4304995    1.805115
        1.web |   -.023855   .4769234    -0.05   0.960     -.959248    .9115381
              |
  web#c.proud |
           1  |   .6393547   .6310294     1.01   0.311    -.5982876    1.876997
              |
  campaignint |   1.532683   .3714496     4.13   0.000     .8041561    2.261209
       polint |   .9761744   .4320923     2.26   0.024     .1287088     1.82364
      polknow |   .5392914   .5038959     1.07   0.285     -.449003    1.527586
       gender |  -.1164197   .1923961    -0.61   0.545    -.4937675    .2609282
        White |  -.5672932   .2113075    -2.68   0.007    -.9817321   -.1528543
       latinx |  -.4355893   .3149479    -1.38   0.167    -1.053299    .1821202
         ageN |   2.976433   .6101052     4.88   0.000     1.779829    4.173036
    education |   1.121319   .3420142     3.28   0.001     .4505241    1.792114
income_norm01 |   1.503712    .354138     4.25   0.000     .8091389    2.198285
        _cons |   -7.14926   .7697912    -9.29   0.000    -8.659057   -5.639463
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file20D")

Average marginal effects

Number of strata =    10                           Number of obs   =     2,575
Number of PSUs   = 1,778                           Population size = 2,291.772
Model VCE: Linearized                              Design df       =     1,768

Expression: Pr(donation), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .0567092   .0454561     1.25   0.212     -.032444    .1458625
          2  |   .1316033   .0373202     3.53   0.000     .0584069    .2047996
------------------------------------------------------------------------------

. 
. combomarginsplot "file20D" "file19D" "file18D" "file17D" "file16D", labels("Donate" "Volunteer" "Display sign" "Attend
>  rally" "Persuade others") recast(scatter) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtit
> le("Marginal Effect of Pride by Mode among Dems") title("") legend(pos(6) row(1)) xsize(8) ysize(6) xlabel(-.6(.2).6) 
> plot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol
> (D)) ci2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(pride12Dem)

  Variables that uniquely identify margins: web _filenumber

. 
. svy: logit persuade c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
                                                   Design df       =     1,379
                                                   F(12, 1368)     =     14.84
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.374513   .4029781     3.41   0.001     .5839963    2.165029
        1.web |  -.0785034   .2273487    -0.35   0.730      -.52449    .3674833
              |
  web#c.proud |
           1  |  -.0370689   .4616971    -0.08   0.936    -.9427735    .8686357
              |
  campaignint |   1.234919     .28126     4.39   0.000     .6831749    1.786663
       polint |   .9211665   .3573222     2.58   0.010     .2202126     1.62212
      polknow |   .6841164    .370529     1.85   0.065    -.0427451    1.410978
       gender |   -.172293    .139406    -1.24   0.217    -.4457639    .1011779
        White |   .6127104   .2631747     2.33   0.020     .0964444    1.128976
       latinx |   .7349745   .3808917     1.93   0.054    -.0122153    1.482164
         ageN |   -.048065   .4839301    -0.10   0.921    -.9973839    .9012539
    education |   .0312533    .266735     0.12   0.907    -.4919968    .5545035
income_norm01 |   .3515714   .2753107     1.28   0.202    -.1885018    .8916445
        _cons |  -3.077518   .4369496    -7.04   0.000    -3.934676    -2.22036
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file16R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
Model VCE: Linearized                              Design df       =     1,379

Expression: Pr(persuade), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .2816387    .073425     3.84   0.000     .1376018    .4256756
          2  |   .2746096   .0467775     5.87   0.000     .1828468    .3663724
------------------------------------------------------------------------------

. svy: logit rally c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      6.55
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .6248918   .7913435     0.79   0.430    -.9274755    2.177259
        1.web |   .1144454   .5052366     0.23   0.821      -.87667    1.105561
              |
  web#c.proud |
           1  |   .2573813    .883072     0.29   0.771    -1.474929    1.989691
              |
  campaignint |   1.002578   .8613417     1.16   0.245    -.6871042    2.692259
       polint |   1.805606   .9861658     1.83   0.067    -.1289418    3.740153
      polknow |   .9864347   .9458566     1.04   0.297    -.8690387    2.841908
       gender |   .3577564   .2712555     1.32   0.187    -.1743616    .8898743
        White |  -.3949455   .4199271    -0.94   0.347    -1.218711    .4288195
       latinx |   .0117897   .7306926     0.02   0.987      -1.4216    1.445179
         ageN |  -.2339589   1.147724    -0.20   0.839    -2.485432    2.017514
    education |   .7060915   .4802526     1.47   0.142    -.2360131    1.648196
income_norm01 |  -.6240241   .5380951    -1.16   0.246    -1.679598    .4315495
        _cons |  -5.877788   .9271692    -6.34   0.000    -7.696603   -4.058974
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file17R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
Model VCE: Linearized                              Design df       =     1,379

Expression: Pr(rally), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .0256778   .0333457     0.77   0.441     -.039736    .0910916
          2  |   .0441138   .0217966     2.02   0.043     .0013558    .0868718
------------------------------------------------------------------------------

. svy: logit button c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyi
> d_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      7.42
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.477868   .5484434     2.69   0.007     .4019941    2.553741
        1.web |   .5508084   .3465669     1.59   0.112    -.1290469    1.230664
              |
  web#c.proud |
           1  |  -.8900297   .6167689    -1.44   0.149    -2.099936     .319877
              |
  campaignint |   1.413899   .4135121     3.42   0.001     .6027184     2.22508
       polint |   1.716884    .509019     3.37   0.001     .7183489     2.71542
      polknow |   -.274043   .5574198    -0.49   0.623    -1.367526    .8194396
       gender |   .1545703   .1915283     0.81   0.420    -.2211481    .5302887
        White |  -.1400815   .3138362    -0.45   0.655    -.7557296    .4755665
       latinx |    -.22335   .5310361    -0.42   0.674    -1.265076    .8183759
         ageN |  -.5892879   .6468149    -0.91   0.362    -1.858135    .6795597
    education |  -.6280902   .3688825    -1.70   0.089    -1.351722    .0955414
income_norm01 |  -.5664775   .3508388    -1.61   0.107    -1.254713     .121758
        _cons |  -3.697182   .6295917    -5.87   0.000    -4.932243   -2.462121
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file18R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
Model VCE: Linearized                              Design df       =     1,379

Expression: Pr(button), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .1470934   .0567439     2.59   0.010     .0357797     .258407
          2  |   .0656988   .0348983     1.88   0.060    -.0027608    .1341584
------------------------------------------------------------------------------

. svy: logit volunteer c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if par
> tyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      5.73
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   7.587754   2.879478     2.64   0.009     1.939122    13.23639
        1.web |    5.22585   2.321585     2.25   0.025     .6716298    9.780069
              |
  web#c.proud |
           1  |  -6.236696   2.934677    -2.13   0.034    -11.99361   -.4797819
              |
  campaignint |   1.022813    .858422     1.19   0.234    -.6611413    2.706767
       polint |   2.367702   1.052717     2.25   0.025     .3026024    4.432801
      polknow |   .2045996   1.184212     0.17   0.863    -2.118453    2.527652
       gender |   .1019049   .3755146     0.27   0.786    -.6347368    .8385466
        White |   -.232624   .5923279    -0.39   0.695    -1.394585    .9293373
       latinx |  -.9216595   .8227725    -1.12   0.263    -2.535681    .6923615
         ageN |  -1.866144   1.691653    -1.10   0.270    -5.184636    1.452347
    education |   .1716704     .55173     0.31   0.756    -.9106505    1.253991
income_norm01 |  -1.974962   .6595057    -2.99   0.003    -3.268705   -.6812195
        _cons |  -10.08137   2.823301    -3.57   0.000     -15.6198   -4.542938
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file19R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
Model VCE: Linearized                              Design df       =     1,379

Expression: Pr(volunteer), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .1113657   .0655083     1.70   0.089    -.0171411    .2398725
          2  |   .0345064   .0164977     2.09   0.037     .0021432    .0668696
------------------------------------------------------------------------------

. svy: logit donation c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      9.03
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.074765   .7211978     1.49   0.136    -.3399981    2.489529
        1.web |   .1495541   .3992137     0.37   0.708    -.6335777    .9326859
              |
  web#c.proud |
           1  |   .0276719   .7833212     0.04   0.972    -1.508958    1.564302
              |
  campaignint |   1.215953   .5600815     2.17   0.030     .1172493    2.314657
       polint |   2.266979   .5593548     4.05   0.000       1.1697    3.364257
      polknow |   1.721819    .655727     2.63   0.009     .4354888    3.008149
       gender |   .1750116   .2113604     0.83   0.408    -.2396111    .5896342
        White |  -.5900464   .3651453    -1.62   0.106    -1.306347    .1262539
       latinx |  -.5403454   .5146406    -1.05   0.294    -1.549908    .4692177
         ageN |   1.025483   .7762632     1.32   0.187    -.4973013    2.548268
    education |   .4615656   .3725064     1.24   0.216    -.2691749    1.192306
income_norm01 |   .6803721   .4177539     1.63   0.104    -.1391298    1.499874
        _cons |   -7.27613    .789305    -9.22   0.000    -8.824498   -5.727762
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file20R")

Average marginal effects

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
Model VCE: Linearized                              Design df       =     1,379

Expression: Pr(donation), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |    .086237   .0605512     1.42   0.155    -.0325454    .2050194
          2  |   .0973809   .0288257     3.38   0.001      .040834    .1539278
------------------------------------------------------------------------------

. 
. combomarginsplot "file20R" "file19R" "file18R" "file17R" "file16R", labels("Donate" "Volunteer" "Display sign" "Attend
>  rally" "Persuade others") recast(scatter) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtit
> le("Marginal Effect of Pride by Mode among Reps") title("") legend(pos(6) row(1)) xsize(8) ysize(6) xlabel(-.6(.2).6) 
> plot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol
> (D)) ci2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(pride12Rep)

  Variables that uniquely identify margins: web _filenumber

. 
. 
. grc1leg anger12Dem anger12Rep fear12Dem fear12Rep hope12Dem hope12Rep pride12Dem pride12Rep, scheme(plottig) xcommon y
> common col(2) imargin(small)

. 
. 
. *********************************APPENDIX****************************************

. 
. *********************

. ** Table B1 -- 2012 **

. *********************

. 
. svy: reg anger web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,759
Number of PSUs   = 1,905                           Population size = 2,455.861
                                                   Design df       =     1,895
                                                   F(10, 1886)     =     14.22
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.0871

-------------------------------------------------------------------------------
              |             Linearized
        anger | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0604935   .0203879     2.97   0.003     .0205085    .1004786
  campaignint |   .2184908   .0340288     6.42   0.000      .151753    .2852287
       polint |   .1284386   .0445211     2.88   0.004     .0411231     .215754
      polknow |  -.0143426   .0513695    -0.28   0.780    -.1150893    .0864042
       gender |    .041748   .0191062     2.19   0.029     .0042767    .0792194
        White |   .0097661   .0231202     0.42   0.673    -.0355776    .0551098
       latinx |  -.0218243   .0309666    -0.70   0.481    -.0825564    .0389079
         ageN |  -.1934399   .0565036    -3.42   0.001    -.3042558   -.0826241
    education |   .0474676   .0370762     1.28   0.201    -.0252469    .1201822
income_norm01 |  -.0241114   .0387397    -0.62   0.534    -.1000884    .0518656
        _cons |   .1861288   .0476253     3.91   0.000     .0927254    .2795322
-------------------------------------------------------------------------------

. outreg2 using tabD12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Anger toward Out
> -Party Candidate") append
tabD12.doc
dir : seeout

. svy: reg fear web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,762
Number of PSUs   = 1,905                           Population size = 2,459.058
                                                   Design df       =     1,895
                                                   F(10, 1886)     =     11.78
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.0664

-------------------------------------------------------------------------------
              |             Linearized
         fear | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |    .052467    .019553     2.68   0.007     .0141194    .0908147
  campaignint |   .1708656   .0311969     5.48   0.000     .1096818    .2320495
       polint |   .1249002   .0451076     2.77   0.006     .0364344    .2133661
      polknow |  -.0681471   .0478613    -1.42   0.155    -.1620134    .0257192
       gender |    .015721   .0185945     0.85   0.398    -.0207468    .0521888
        White |   .0139536   .0223959     0.62   0.533    -.0299696    .0578768
       latinx |   -.067328   .0273262    -2.46   0.014    -.1209207   -.0137353
         ageN |  -.0367766   .0558098    -0.66   0.510    -.1462317    .0726785
    education |   .0280592   .0347867     0.81   0.420    -.0401651    .0962835
income_norm01 |    .015249   .0388116     0.39   0.694     -.060869     .091367
        _cons |   .1091621   .0447696     2.44   0.015     .0213593     .196965
-------------------------------------------------------------------------------

. outreg2 using tabD12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Fear toward Out-
> Party Candidate") append
tabD12.doc
dir : seeout

. svy: reg hopeful web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,768
Number of PSUs   = 1,903                           Population size = 2,463.194
                                                   Design df       =     1,893
                                                   F(10, 1884)     =     21.73
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1397

-------------------------------------------------------------------------------
              |             Linearized
      hopeful | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |  -.0564577   .0182821    -3.09   0.002    -.0923129   -.0206024
  campaignint |   .1655221   .0273455     6.05   0.000     .1118916    .2191527
       polint |   .1624914   .0353207     4.60   0.000     .0932198     .231763
      polknow |   .0114417   .0401681     0.28   0.776    -.0673368    .0902201
       gender |   .0423603   .0157956     2.68   0.007     .0113818    .0733389
        White |  -.1379781   .0190662    -7.24   0.000     -.175371   -.1005852
       latinx |  -.0642323   .0255115    -2.52   0.012    -.1142659   -.0141987
         ageN |  -.0167868   .0458604    -0.37   0.714     -.106729    .0731554
    education |   .0119494   .0312784     0.38   0.702    -.0493943    .0732932
income_norm01 |   .0065117   .0318737     0.20   0.838    -.0559996    .0690231
        _cons |   .4177186   .0432538     9.66   0.000     .3328883    .5025488
-------------------------------------------------------------------------------

. outreg2 using tabD12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Hope toward In-P
> arty Candidate") append
tabD12.doc
dir : seeout

. svy: reg proud web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,763
Number of PSUs   = 1,903                           Population size = 2,459.566
                                                   Design df       =     1,893
                                                   F(10, 1884)     =     38.38
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1940

-------------------------------------------------------------------------------
              |             Linearized
        proud | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |  -.0549685   .0197912    -2.78   0.006    -.0937834   -.0161536
  campaignint |   .2072205   .0298975     6.93   0.000      .148585    .2658561
       polint |   .1934217   .0377299     5.13   0.000     .1194251    .2674182
      polknow |  -.0195274   .0415734    -0.47   0.639     -.101062    .0620072
       gender |    .045237    .016654     2.72   0.007     .0125749     .077899
        White |  -.2100214   .0201465   -10.42   0.000    -.2495332   -.1705097
       latinx |  -.1556803   .0267129    -5.83   0.000    -.2080701   -.1032905
         ageN |   .0519563   .0504416     1.03   0.303    -.0469707    .1508832
    education |  -.0036547   .0313537    -0.12   0.907    -.0651462    .0578368
income_norm01 |   .0286753   .0322089     0.89   0.373    -.0344934    .0918439
        _cons |   .4024261   .0434749     9.26   0.000     .3171622    .4876899
-------------------------------------------------------------------------------

. outreg2 using tabD12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Pride toward In-
> Party Candidate") append
tabD12.doc
dir : seeout

. 
. *********************

. ** Table B2 -- 2012 **

. *********************

. 
. svy: reg anger web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,836
Number of PSUs   = 1,489                           Population size = 2,121.787
                                                   Design df       =     1,479
                                                   F(10, 1470)     =     25.74
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1352

-------------------------------------------------------------------------------
              |             Linearized
        anger | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0807825   .0193493     4.17   0.000     .0428275    .1187374
  campaignint |   .1687951   .0367129     4.60   0.000     .0967802      .24081
       polint |    .208203   .0496239     4.20   0.000     .1108623    .3055437
      polknow |  -.0149873   .0522918    -0.29   0.774    -.1175612    .0875867
       gender |  -.0050757   .0175449    -0.29   0.772    -.0394912    .0293399
        White |   .0948241   .0431216     2.20   0.028     .0102381      .17941
       latinx |   .0335752   .0548915     0.61   0.541    -.0740983    .1412487
         ageN |   .1017905   .0604643     1.68   0.092    -.0168144    .2203954
    education |  -.0659144   .0368009    -1.79   0.073    -.1381019    .0062731
income_norm01 |    .020241   .0356664     0.57   0.570     -.049721    .0902031
        _cons |    .066775   .0563613     1.18   0.236    -.0437816    .1773316
-------------------------------------------------------------------------------

. outreg2 using tabR12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Anger toward Out
> -Party Candidate") append
tabR12.doc
dir : seeout

. svy: reg fear web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,838
Number of PSUs   = 1,490                           Population size = 2,124.153
                                                   Design df       =     1,480
                                                   F(10, 1471)     =     25.30
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1237

-------------------------------------------------------------------------------
              |             Linearized
         fear | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .1236523   .0199834     6.19   0.000     .0844536     .162851
  campaignint |   .1603995   .0379182     4.23   0.000     .0860204    .2347785
       polint |   .1569721   .0494941     3.17   0.002      .059886    .2540583
      polknow |    .039603    .053042     0.75   0.455    -.0644425    .1436485
       gender |   .0343117   .0194745     1.76   0.078    -.0038888    .0725121
        White |   .0785247   .0422424     1.86   0.063    -.0043366    .1613859
       latinx |   .0170786   .0562263     0.30   0.761    -.0932131    .1273704
         ageN |   .1544932   .0604455     2.56   0.011     .0359252    .2730612
    education |  -.0618392   .0403049    -1.53   0.125    -.1409001    .0172216
income_norm01 |   .0095667   .0381423     0.25   0.802     -.065252    .0843854
        _cons |  -.0857731   .0543123    -1.58   0.114    -.1923104    .0207641
-------------------------------------------------------------------------------

. outreg2 using tabR12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Fear toward Out-
> Party Candidate") append
tabR12.doc
dir : seeout

. svy: reg hopeful web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,832
Number of PSUs   = 1,489                           Population size = 2,117.583
                                                   Design df       =     1,479
                                                   F(10, 1470)     =     22.66
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1580

-------------------------------------------------------------------------------
              |             Linearized
      hopeful | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0097969   .0210202     0.47   0.641    -.0314357    .0510294
  campaignint |   .2169213    .036768     5.90   0.000     .1447982    .2890443
       polint |   .1796436   .0485619     3.70   0.000     .0843861    .2749011
      polknow |   .0146573   .0491513     0.30   0.766    -.0817564    .1110709
       gender |   .0635293   .0175517     3.62   0.000     .0291004    .0979582
        White |   .0558678   .0475416     1.18   0.240    -.0373883    .1491239
       latinx |   .0761422   .0615633     1.24   0.216    -.0446184    .1969028
         ageN |   .1429005   .0613606     2.33   0.020     .0225374    .2632637
    education |  -.0763444   .0382885    -1.99   0.046      -.15145   -.0012388
income_norm01 |   .0720697   .0350025     2.06   0.040       .00341    .1407294
        _cons |   .0324221    .069825     0.46   0.642    -.1045445    .1693887
-------------------------------------------------------------------------------

. outreg2 using tabR12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Hope toward In-P
> arty Candidate") append
tabR12.doc
dir : seeout

. svy: reg proud web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,823
Number of PSUs   = 1,490                           Population size = 2,101.327
                                                   Design df       =     1,480
                                                   F(10, 1471)     =     32.77
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1760

-------------------------------------------------------------------------------
              |             Linearized
        proud | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0474718    .023509     2.02   0.044     .0013573    .0935863
  campaignint |   .2345256   .0394658     5.94   0.000     .1571108    .3119404
       polint |   .2043627   .0511975     3.99   0.000     .1039352    .3047901
      polknow |  -.0252561   .0549819    -0.46   0.646    -.1331068    .0825947
       gender |   .1120389   .0191052     5.86   0.000     .0745628    .1495151
        White |   .0689706   .0419219     1.65   0.100    -.0132621    .1512033
       latinx |   .0742721   .0562438     1.32   0.187    -.0360539    .1845981
         ageN |   .1718569   .0627263     2.74   0.006      .048815    .2948988
    education |  -.0409834   .0382022    -1.07   0.284    -.1159197    .0339529
income_norm01 |   .0068304   .0349014     0.20   0.845     -.061631    .0752918
        _cons |  -.1502946   .0557804    -2.69   0.007    -.2597116   -.0408775
-------------------------------------------------------------------------------

. outreg2 using tabR12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Pride toward In-
> Party Candidate") append
tabR12.doc
dir : seeout

. 
. *********************

. ** Table B5 -- 2012 **

. *********************

. 
. svy: logit persuade c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,288.471
                                                   Design df       =     1,770
                                                   F(12, 1759)     =     15.08
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.000948   .2849785     3.51   0.000     .4420183    1.559878
        1.web |   .1778652    .206506     0.86   0.389    -.2271561    .5828864
              |
  web#c.anger |
           1  |  -.3669051   .3463839    -1.06   0.290     -1.04627    .3124595
              |
  campaignint |    1.48007   .2369267     6.25   0.000     1.015385    1.944756
       polint |   1.332827   .2869763     4.64   0.000     .7699786    1.895675
      polknow |  -.2137451   .3153266    -0.68   0.498    -.8321968    .4047065
       gender |  -.0704108   .1257511    -0.56   0.576    -.3170472    .1762255
        White |  -.2852514   .1478975    -1.93   0.054    -.5753235    .0048208
       latinx |  -.3351057   .2022054    -1.66   0.098    -.7316922    .0614809
         ageN |   .4554457   .3961999     1.15   0.250    -.3216232    1.232515
    education |  -.3754531   .2369555    -1.58   0.113    -.8401952     .089289
income_norm01 |   .4719472   .2366366     1.99   0.046     .0078307    .9360638
        _cons |  -2.553599   .3252943    -7.85   0.000      -3.1916   -1.915598
-------------------------------------------------------------------------------

. outreg2 using tabDemanger12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemanger12.doc
dir : seeout

. svy: logit rally c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,572
Number of PSUs   = 1,780                           Population size = 2,288.169
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      7.20
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .4898116    .447532     1.09   0.274    -.3879353    1.367558
        1.web |   .0323276   .4199736     0.08   0.939    -.7913688     .856024
              |
  web#c.anger |
           1  |  -.0184238   .5952964    -0.03   0.975    -1.185982    1.149134
              |
  campaignint |   2.006626   .4513077     4.45   0.000     1.121474    2.891778
       polint |   .7411165   .5712955     1.30   0.195    -.3793683    1.861601
      polknow |   .0241461   .5705061     0.04   0.966    -1.094791    1.143083
       gender |  -.0725974    .249202    -0.29   0.771    -.5613585    .4161637
        White |   -.679304   .2096912    -3.24   0.001    -1.090572   -.2680356
       latinx |  -1.176752   .4520148    -2.60   0.009    -2.063291   -.2902133
         ageN |   .4346705   .7976541     0.54   0.586    -1.129773    1.999114
    education |    1.29451   .4444615     2.91   0.004     .4227853    2.166235
income_norm01 |  -.2473522   .4198194    -0.59   0.556    -1.070746    .5760417
        _cons |  -5.194451   .9375082    -5.54   0.000    -7.033191   -3.355711
-------------------------------------------------------------------------------

. outreg2 using tabDemanger12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemanger12.doc
dir : seeout

. svy: logit button c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyi
> d_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,288.471
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      8.62
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .9910783   .3223973     3.07   0.002     .3587589    1.623398
        1.web |   .4348459   .2697735     1.61   0.107    -.0942623     .963954
              |
  web#c.anger |
           1  |  -.1853941   .4175661    -0.44   0.657    -1.004369    .6335803
              |
  campaignint |   1.532672   .3106589     4.93   0.000      .923375    2.141969
       polint |   .4409262   .3463474     1.27   0.203    -.2383667    1.120219
      polknow |   -.113183   .3629758    -0.31   0.755    -.8250894    .5987233
       gender |  -.0260055   .1618597    -0.16   0.872    -.3434618    .2914508
        White |  -.9367899   .1630407    -5.75   0.000    -1.256563   -.6170173
       latinx |  -.7571445   .2305721    -3.28   0.001    -1.209367   -.3049222
         ageN |   .6723597   .4670242     1.44   0.150    -.2436173    1.588337
    education |  -.4837513   .2769574    -1.75   0.081    -1.026949    .0594467
income_norm01 |   .2966654   .2728786     1.09   0.277    -.2385328    .8318635
        _cons |  -3.091563   .4709028    -6.57   0.000    -4.015147   -2.167979
-------------------------------------------------------------------------------

. outreg2 using tabDemanger12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemanger12.doc
dir : seeout

. svy: logit volunteer c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if par
> tyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,288.471
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      5.74
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.245106   .5748576     2.17   0.030     .1176351    2.372577
        1.web |    .186315   .4724389     0.39   0.693    -.7402818    1.112912
              |
  web#c.anger |
           1  |  -.3081413   .7248005    -0.43   0.671    -1.729696    1.113414
              |
  campaignint |   1.421751   .5743389     2.48   0.013     .2952972    2.548205
       polint |   1.266233   .7157865     1.77   0.077    -.1376428    2.670109
      polknow |   .0214423   .6667188     0.03   0.974    -1.286197    1.329081
       gender |  -.1348868    .290096    -0.46   0.642    -.7038535    .4340799
        White |  -.5912431   .2787647    -2.12   0.034    -1.137986   -.0445003
       latinx |  -.5203619   .4801858    -1.08   0.279    -1.462153    .4214289
         ageN |   1.931107   .9674569     2.00   0.046     .0336287    3.828585
    education |   .7552936    .469118     1.61   0.108    -.1647899    1.675377
income_norm01 |   .2527119   .4362437     0.58   0.562    -.6028952    1.108319
        _cons |  -6.633148   1.189095    -5.58   0.000    -8.965326   -4.300969
-------------------------------------------------------------------------------

. outreg2 using tabDemanger12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemanger12.doc
dir : seeout

. svy: logit donation c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,571
Number of PSUs   = 1,780                           Population size = 2,287.711
                                                   Design df       =     1,770
                                                   F(12, 1759)     =     10.93
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .8101405   .4380256     1.85   0.065    -.0489614    1.669242
        1.web |     .08887   .3324687     0.27   0.789    -.5632026    .7409426
              |
  web#c.anger |
           1  |   .4634777   .5410634     0.86   0.392    -.5977126    1.524668
              |
  campaignint |   1.468052   .3490984     4.21   0.000     .7833639    2.152741
       polint |   1.022937   .4248781     2.41   0.016     .1896219    1.856253
      polknow |   .5309804   .4984081     1.07   0.287    -.4465499    1.508511
       gender |  -.0974029    .189597    -0.51   0.608    -.4692605    .2744548
        White |  -.7850236    .201773    -3.89   0.000    -1.180762   -.3892852
       latinx |  -.5126506   .3108013    -1.65   0.099    -1.122227    .0969256
         ageN |   3.286107   .6225569     5.28   0.000     2.065083    4.507131
    education |    1.04503   .3324316     3.14   0.002     .3930306     1.69703
income_norm01 |   1.597129   .3556278     4.49   0.000     .8996348    2.294624
        _cons |  -7.090438   .7736289    -9.17   0.000     -8.60776   -5.573116
-------------------------------------------------------------------------------

. outreg2 using tabDemanger12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemanger12.doc
dir : seeout

. 
. *********************

. ** Table B6 -- 2012 **

. *********************

. 
. svy: logit persuade c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
                                                   Design df       =     1,378
                                                   F(12, 1367)     =     15.67
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.104662   .4782444     2.31   0.021     .1664959    2.042828
        1.web |  -.4209978   .2843176    -1.48   0.139      -.97874    .1367444
              |
  web#c.anger |
           1  |   .5445071   .5384748     1.01   0.312    -.5118119    1.600826
              |
  campaignint |   1.240042    .279547     4.44   0.000      .691658    1.788425
       polint |   .9333729   .3751283     2.49   0.013     .1974886    1.669257
      polknow |   .7812633   .3704113     2.11   0.035     .0546323    1.507894
       gender |  -.0101028   .1413929    -0.07   0.943    -.2874713    .2672658
        White |    .593357   .2566475     2.31   0.021     .0898948    1.096819
       latinx |   .8204393   .3763994     2.18   0.029     .0820614    1.558817
         ageN |   .0981278   .4816874     0.20   0.839    -.8467921    1.043048
    education |   -.007393   .2726387    -0.03   0.978    -.5422247    .5274388
income_norm01 |   .3783991    .279617     1.35   0.176     -.170122    .9269202
        _cons |  -3.306137   .4628174    -7.14   0.000     -4.21404   -2.398234
-------------------------------------------------------------------------------

. outreg2 using tabRepanger12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepanger12.doc
dir : seeout

. svy: logit rally c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      6.60
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .5689928   .9131699     0.62   0.533    -1.222361    2.360346
        1.web |  -.0785164   .5993188    -0.13   0.896    -1.254192    1.097159
              |
  web#c.anger |
           1  |   .1527747   1.036642     0.15   0.883    -1.880792    2.186342
              |
  campaignint |   .9413843   .8643409     1.09   0.276    -.7541819    2.636951
       polint |   1.905424   .9158462     2.08   0.038     .1088209    3.702028
      polknow |   .8960942   .9155447     0.98   0.328    -.8999179    2.692106
       gender |   .3545847   .2574662     1.38   0.169    -.1504833    .8596527
        White |  -.3251093   .4214698    -0.77   0.441    -1.151901    .5016826
       latinx |   .1615584   .7202463     0.22   0.823    -1.251339    1.574456
         ageN |     .39337   1.128675     0.35   0.728    -1.820736    2.607476
    education |   .8729293   .4701263     1.86   0.064    -.0493114     1.79517
income_norm01 |  -.7438118   .5119954    -1.45   0.147    -1.748187     .260563
        _cons |  -6.077841   1.007764    -6.03   0.000    -8.054757   -4.100924
-------------------------------------------------------------------------------

. outreg2 using tabRepanger12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepanger12.doc
dir : seeout

. svy: logit button c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyi
> d_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      8.67
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.509399   .6007244     2.51   0.012     .3309656    2.687832
        1.web |   .5184399   .4253077     1.22   0.223    -.3158808    1.352761
              |
  web#c.anger |
           1  |  -.8941779   .6785599    -1.32   0.188      -2.2253    .4369443
              |
  campaignint |   1.332825   .4251186     3.14   0.002     .4988755    2.166775
       polint |   1.840732   .5139273     3.58   0.000     .8325677    2.848897
      polknow |  -.2807667   .5488841    -0.51   0.609    -1.357505     .795972
       gender |   .2442395   .1835598     1.33   0.184    -.1158474    .6043264
        White |  -.0865676   .3238022    -0.27   0.789    -.7217661    .5486309
       latinx |   -.085823    .524415    -0.16   0.870    -1.114561    .9429151
         ageN |  -.2226635   .6372925    -0.35   0.727    -1.472832    1.027505
    education |  -.4345283    .374269    -1.16   0.246    -1.168727    .2996702
income_norm01 |  -.6541931   .3509891    -1.86   0.063    -1.342724    .0343376
        _cons |  -4.113432   .6675685    -6.16   0.000    -5.422992   -2.803871
-------------------------------------------------------------------------------

. outreg2 using tabRepanger12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepanger12.doc
dir : seeout

. svy: logit volunteer c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if par
> tyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      5.72
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.023475   1.293739     0.79   0.429    -1.514437    3.561386
        1.web |   .4688513   1.005084     0.47   0.641    -1.502809    2.440511
              |
  web#c.anger |
           1  |  -.5296643   1.529271    -0.35   0.729    -3.529615    2.470287
              |
  campaignint |   .9176879   .8071331     1.14   0.256    -.6656545     2.50103
       polint |   2.677664   .9308742     2.88   0.004     .8515801    4.503748
      polknow |  -.0703519   1.291189    -0.05   0.957    -2.603261    2.462557
       gender |  -.0259289   .4095092    -0.06   0.950    -.8292579       .7774
        White |  -.1144153   .5606749    -0.20   0.838    -1.214284    .9854534
       latinx |  -.5231716   .7509117    -0.70   0.486    -1.996225    .9498821
         ageN |  -.0990947   1.769158    -0.06   0.955    -3.569628    3.371439
    education |   .2793235    .540336     0.52   0.605    -.7806466    1.339294
income_norm01 |  -2.123718   .6238833    -3.40   0.001    -3.347582   -.8998545
        _cons |   -5.85565    1.43698    -4.07   0.000    -8.674555   -3.036745
-------------------------------------------------------------------------------

. outreg2 using tabRepanger12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepanger12.doc
dir : seeout

. svy: logit donation c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,709
Number of PSUs   = 1,388                           Population size = 1,977.687
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      9.04
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.051457   .7737883     1.36   0.174    -.4664735    2.569387
        1.web |   .1820186   .5010553     0.36   0.716    -.8008951    1.164932
              |
  web#c.anger |
           1  |  -.1692108   .8410452    -0.20   0.841    -1.819078    1.480657
              |
  campaignint |   1.245248   .5561912     2.24   0.025      .154175    2.336321
       polint |   2.419008    .554988     4.36   0.000     1.330295    3.507721
      polknow |   1.812027   .6799838     2.66   0.008     .4781116    3.145942
       gender |   .2872639   .2125904     1.35   0.177    -.1297718    .7042996
        White |  -.5432975   .3622597    -1.50   0.134    -1.253938    .1673426
       latinx |  -.4064512   .5523007    -0.74   0.462    -1.489892    .6769899
         ageN |   1.243527   .7923763     1.57   0.117    -.3108669    2.797922
    education |    .486355   .3730979     1.30   0.193    -.2455462    1.218256
income_norm01 |   .6554249   .4092824     1.60   0.110    -.1474591    1.458309
        _cons |  -7.746195   .9233856    -8.39   0.000    -9.557588   -5.934801
-------------------------------------------------------------------------------

. outreg2 using tabRepanger12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepanger12.doc
dir : seeout

. 
. *********************

. ** Table B7 -- 2012 **

. *********************

. 
. svy: logit persuade c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,574
Number of PSUs   = 1,780                           Population size = 2,286.622
                                                   Design df       =     1,770
                                                   F(12, 1759)     =     15.32
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.150689   .2921312     3.94   0.000     .5777304    1.723647
        1.web |   .1786673    .183769     0.97   0.331    -.1817598    .5390945
              |
   web#c.fear |
           1  |   -.467355    .351627    -1.33   0.184    -1.157003    .2222929
              |
  campaignint |    1.51767   .2335696     6.50   0.000     1.059568    1.975771
       polint |   1.343466   .2846857     4.72   0.000     .7851106    1.901822
      polknow |   -.193451     .31221    -0.62   0.536      -.80579    .4188881
       gender |  -.0561052   .1246556    -0.45   0.653    -.3005929    .1883825
        White |  -.2954651   .1484088    -1.99   0.047    -.5865401   -.0043901
       latinx |  -.3160159   .2040973    -1.55   0.122    -.7163131    .0842813
         ageN |   .3455646   .3904831     0.88   0.376     -.420292    1.111421
    education |  -.3688208   .2363175    -1.56   0.119    -.8323115    .0946699
income_norm01 |   .4436726   .2356187     1.88   0.060    -.0184476    .9057927
        _cons |  -2.504448    .305797    -8.19   0.000    -3.104209   -1.904686
-------------------------------------------------------------------------------

. outreg2 using tabDemfear12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemfear12.doc
dir : seeout

. svy: logit rally c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_
> 2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,574
Number of PSUs   = 1,780                           Population size = 2,291.233
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      5.80
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .6510495   .4827514     1.35   0.178    -.2957734    1.597872
        1.web |  -.0034542   .3343612    -0.01   0.992    -.6592386    .6523301
              |
   web#c.fear |
           1  |  -.2300797   .5875166    -0.39   0.695    -1.382379    .9222195
              |
  campaignint |   1.481003   .6285218     2.36   0.019     .2482801    2.713726
       polint |   .9601058    .602521     1.59   0.111    -.2216217    2.141833
      polknow |   .0383597   .5609811     0.07   0.945    -1.061895    1.138615
       gender |  -.1426372    .249584    -0.57   0.568    -.6321476    .3468732
        White |  -.6153872   .2146969    -2.87   0.004    -1.036473   -.1943011
       latinx |  -1.228348   .4532981    -2.71   0.007    -2.117404   -.3392922
         ageN |  -.0554976   .8534717    -0.07   0.948    -1.729416    1.618421
    education |   .9973306   .5018848     1.99   0.047     .0129814     1.98168
income_norm01 |   .0089145   .4704631     0.02   0.985    -.9138072    .9316363
        _cons |  -4.533858   .9814305    -4.62   0.000    -6.458742   -2.608973
-------------------------------------------------------------------------------

. outreg2 using tabDemfear12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemfear12.doc
dir : seeout

. svy: logit button c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,575
Number of PSUs   = 1,780                           Population size = 2,291.535
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      9.20
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .8718486   .3377918     2.58   0.010     .2093358    1.534361
        1.web |   .2379301   .2382932     1.00   0.318    -.2294355    .7052957
              |
   web#c.fear |
           1  |   .0884316   .4231199     0.21   0.834    -.7414357    .9182989
              |
  campaignint |   1.398541   .3408679     4.10   0.000     .7299954    2.067087
       polint |   .5506537   .3523762     1.56   0.118    -.1404635    1.241771
      polknow |  -.0856225   .3623448    -0.24   0.813    -.7962912    .6250462
       gender |  -.0417525   .1616437    -0.26   0.796    -.3587852    .2752803
        White |  -.9347788   .1661845    -5.62   0.000    -1.260717   -.6088402
       latinx |  -.7433261    .229689    -3.24   0.001    -1.193816   -.2928359
         ageN |    .392118   .4938092     0.79   0.427    -.5763926    1.360629
    education |  -.5830938   .2958492    -1.97   0.049    -1.163344   -.0028433
income_norm01 |   .3518379   .2823273     1.25   0.213    -.2018921    .9055678
        _cons |  -2.709578   .4761507    -5.69   0.000    -3.643455   -1.775701
-------------------------------------------------------------------------------

. outreg2 using tabDemfear12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemfear12.doc
dir : seeout

. svy: logit volunteer c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,575
Number of PSUs   = 1,780                           Population size = 2,291.535
                                                   Design df       =     1,770
                                                   F(12, 1759)     =      5.65
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .8160819   .5925434     1.38   0.169    -.3460765     1.97824
        1.web |   .1198022    .405616     0.30   0.768    -.6757345     .915339
              |
   web#c.fear |
           1  |  -.2597675   .7265523    -0.36   0.721    -1.684758    1.165223
              |
  campaignint |   1.570046   .5853029     2.68   0.007      .422088    2.718003
       polint |   1.364655   .7148476     1.91   0.056     -.037379     2.76669
      polknow |  -.0331229   .6709975    -0.05   0.961    -1.349154    1.282908
       gender |  -.1109767   .2873475    -0.39   0.699    -.6745528    .4525994
        White |   -.598726    .275805    -2.17   0.030    -1.139664   -.0577882
       latinx |  -.5485716   .4732485    -1.16   0.247    -1.476756    .3796132
         ageN |   1.822006   .9458556     1.93   0.054    -.0331052    3.677118
    education |   .7512863   .4636122     1.62   0.105    -.1579986    1.660571
income_norm01 |   .2369363   .4299802     0.55   0.582     -.606386    1.080259
        _cons |  -6.421128   1.172456    -5.48   0.000    -8.720672   -4.121585
-------------------------------------------------------------------------------

. outreg2 using tabDemfear12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemfear12.doc
dir : seeout

. svy: logit donation c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,573
Number of PSUs   = 1,780                           Population size = 2,290.775
                                                   Design df       =     1,770
                                                   F(12, 1759)     =     12.67
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.231789   .4140905     2.97   0.003      .419631    2.043947
        1.web |   .3589385   .3037741     1.18   0.238    -.2368552    .9547321
              |
   web#c.fear |
           1  |  -.1372147   .5102512    -0.27   0.788    -1.137973    .8635437
              |
  campaignint |   1.502305    .354017     4.24   0.000     .8079691     2.19664
       polint |   1.072888    .436224     2.46   0.014     .2173198    1.928456
      polknow |   .5779862   .5133003     1.13   0.260    -.4287524    1.584725
       gender |  -.1000945   .1899204    -0.53   0.598    -.4725863    .2723973
        White |  -.7753275   .2034941    -3.81   0.000    -1.174442   -.3762134
       latinx |  -.4832293      .3093    -1.56   0.118    -1.089861    .1234024
         ageN |    3.12489   .6174924     5.06   0.000     1.913799    4.335981
    education |   1.027634   .3319533     3.10   0.002     .3765724    1.678696
income_norm01 |   1.603158   .3453902     4.64   0.000     .9257423    2.280573
        _cons |  -7.195305   .7256746    -9.92   0.000    -8.618574   -5.772035
-------------------------------------------------------------------------------

. outreg2 using tabDemfear12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemfear12.doc
dir : seeout

. 
. *********************

. ** Table B8 -- 2012 **

. *********************

. 
. svy: logit persuade c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
                                                   Design df       =     1,379
                                                   F(12, 1368)     =     14.43
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .6737774   .4515579     1.49   0.136    -.2120374    1.559592
        1.web |  -.4411361   .2346545    -1.88   0.060    -.9014546    .0191824
              |
   web#c.fear |
           1  |   .7811883   .5058279     1.54   0.123     -.211087    1.773464
              |
  campaignint |   1.308765   .2811618     4.65   0.000     .7572136    1.860316
       polint |    1.00852   .3661111     2.75   0.006     .2903248    1.726715
      polknow |   .6396292    .377596     1.69   0.091    -.1010956    1.380354
       gender |  -.0525927   .1402368    -0.38   0.708    -.3276933    .2225078
        White |   .6171972    .253829     2.43   0.015     .1192645     1.11513
       latinx |   .8490518   .3921001     2.17   0.031     .0798746    1.618229
         ageN |   .0405783   .4914826     0.08   0.934    -.9235562    1.004713
    education |  -.0304296    .270976    -0.11   0.911    -.5619993    .5011401
income_norm01 |   .3889073   .2734765     1.42   0.155    -.1475677    .9253823
        _cons |  -3.021267    .429024    -7.04   0.000    -3.862877   -2.179656
-------------------------------------------------------------------------------

. outreg2 using tabRepfear12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepfear12.doc
dir : seeout

. svy: logit rally c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_
> 2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      6.47
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.293185   .7225721     1.79   0.074    -.1242749    2.710644
        1.web |   .5259651   .4919089     1.07   0.285    -.4390057    1.490936
              |
   web#c.fear |
           1  |  -1.178744   .8580556    -1.37   0.170     -2.86198    .5044911
              |
  campaignint |   1.021426   .8262964     1.24   0.217    -.5995082     2.64236
       polint |   1.952373   .9051437     2.16   0.031     .1767652     3.72798
      polknow |    .950974   .9016614     1.05   0.292    -.8178023     2.71975
       gender |   .3440257   .2704806     1.27   0.204    -.1865723    .8746237
        White |  -.2821064   .4254635    -0.66   0.507    -1.116732    .5525193
       latinx |   .1912371   .7204026     0.27   0.791    -1.221966    1.604441
         ageN |   .3610532   1.130449     0.32   0.749    -1.856533     2.57864
    education |   .8710918   .4872718     1.79   0.074    -.0847823    1.826966
income_norm01 |  -.7484524   .5133034    -1.46   0.145    -1.755392    .2584875
        _cons |  -6.472773   .9455745    -6.85   0.000    -8.327693   -4.617853
-------------------------------------------------------------------------------

. outreg2 using tabRepfear12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepfear12.doc
dir : seeout

. svy: logit button c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      8.64
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.364539    .447409     3.05   0.002     .4868629    2.242215
        1.web |   .3529496   .3152251     1.12   0.263     -.265423    .9713222
              |
   web#c.fear |
           1  |  -.8326974   .5336994    -1.56   0.119    -1.879648    .2142531
              |
  campaignint |   1.391341   .4039565     3.44   0.001     .5989055    2.183777
       polint |    1.85852   .5055731     3.68   0.000     .8667446    2.850296
      polknow |  -.3403807    .541649    -0.63   0.530    -1.402926    .7221643
       gender |   .1928221   .1840572     1.05   0.295    -.1682403    .5538845
        White |  -.0760733   .3152496    -0.24   0.809     -.694494    .5423475
       latinx |  -.0805887   .5287278    -0.15   0.879    -1.117787    .9566092
         ageN |  -.2373279    .630532    -0.38   0.707    -1.474234    .9995778
    education |   -.439941   .3726177    -1.18   0.238      -1.1709    .2910178
income_norm01 |  -.6724265   .3525821    -1.91   0.057    -1.364082    .0192287
        _cons |  -3.838898   .6288496    -6.10   0.000    -5.072503   -2.605292
-------------------------------------------------------------------------------

. outreg2 using tabRepfear12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepfear12.doc
dir : seeout

. svy: logit volunteer c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      5.14
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.253707   1.492212     0.84   0.401    -1.673544    4.180958
        1.web |   .0166701   .9663474     0.02   0.986       -1.879     1.91234
              |
   web#c.fear |
           1  |   .0412075   1.589338     0.03   0.979    -3.076575     3.15899
              |
  campaignint |   .8088518   .7858562     1.03   0.304    -.7327512    2.350455
       polint |   2.659002   .9821166     2.71   0.007     .7323977    4.585606
      polknow |  -.2505869   1.325876    -0.19   0.850    -2.851538    2.350364
       gender |  -.1066345   .4244404    -0.25   0.802    -.9392532    .7259841
        White |  -.0787657   .5943932    -0.13   0.895    -1.244778    1.087247
       latinx |  -.4565339   .7875244    -0.58   0.562    -2.001409    1.088342
         ageN |   -.256133   1.787047    -0.14   0.886    -3.761757    3.249491
    education |   .3634774   .5436772     0.67   0.504    -.7030464    1.430001
income_norm01 |  -2.223438   .6362215    -3.49   0.000    -3.471505   -.9753714
        _cons |  -5.494324   1.496663    -3.67   0.000    -8.430307   -2.558342
-------------------------------------------------------------------------------

. outreg2 using tabRepfear12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepfear12.doc
dir : seeout

. svy: logit donation c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,711
Number of PSUs   = 1,389                           Population size = 1,980.053
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      9.11
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .5077391   .5869131     0.87   0.387    -.6436001    1.659078
        1.web |  -.0317586   .3559731    -0.09   0.929    -.7300659    .6665488
              |
   web#c.fear |
           1  |   .2470781   .6483762     0.38   0.703    -1.024832    1.518988
              |
  campaignint |   1.335338   .5396922     2.47   0.013     .2766316    2.394044
       polint |    2.46245   .5547592     4.44   0.000     1.374186    3.550713
      polknow |   1.738384   .6810018     2.55   0.011     .4024724    3.074295
       gender |   .2428156    .212441     1.14   0.253    -.1739268     .659558
        White |  -.5560727   .3607539    -1.54   0.123    -1.263759    .1516131
       latinx |  -.4216921   .5384011    -0.78   0.434    -1.477866    .6344816
         ageN |   1.256605   .7850037     1.60   0.110    -.2833253    2.796536
    education |   .4726119   .3656721     1.29   0.196    -.2447218    1.189946
income_norm01 |    .667887   .4136091     1.61   0.107    -.1434841    1.479258
        _cons |  -7.426187   .8062679    -9.21   0.000    -9.007831   -5.844543
-------------------------------------------------------------------------------

. outreg2 using tabRepfear12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepfear12.doc
dir : seeout

. 
. *********************

. ** Table B9 -- 2012 **

. *********************

. 
. svy: logit persuade c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if pa
> rtyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,580
Number of PSUs   = 1,778                           Population size = 2,290.758
                                                   Design df       =     1,768
                                                   F(12, 1757)     =     13.68
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   .1030272   .3848212     0.27   0.789    -.6517251    .8577796
        1.web |  -.0108515   .2862426    -0.04   0.970     -.572261     .550558
              |
web#c.hopeful |
           1  |   .1856627   .4458197     0.42   0.677    -.6887264    1.060052
              |
  campaignint |    1.59667   .2402241     6.65   0.000     1.125517    2.067823
       polint |   1.401529    .286492     4.89   0.000       .83963    1.963427
      polknow |  -.2499356   .3170882    -0.79   0.431    -.8718427    .3719716
       gender |  -.0484427   .1250961    -0.39   0.699    -.2937945    .1969092
        White |  -.2519815   .1507255    -1.67   0.095    -.5476004    .0436373
       latinx |  -.3487115   .2003353    -1.74   0.082    -.7416305    .0442075
         ageN |   .3544806   .3917601     0.90   0.366    -.4138811    1.122842
    education |  -.3212425   .2371419    -1.35   0.176    -.7863504    .1438654
income_norm01 |   .4319908   .2398324     1.80   0.072    -.0383942    .9023757
        _cons |  -2.343624   .3628109    -6.46   0.000    -3.055208   -1.632041
-------------------------------------------------------------------------------

. outreg2 using tabDemhope12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemhope12.doc
dir : seeout

. svy: logit rally c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,580
Number of PSUs   = 1,778                           Population size = 2,295.369
                                                   Design df       =     1,768
                                                   F(12, 1757)     =      5.40
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |    .886592   .9125824     0.97   0.331    -.9032619    2.676446
        1.web |   .1027892   .7843816     0.13   0.896    -1.435624    1.641202
              |
web#c.hopeful |
           1  |   -.207837   1.037355    -0.20   0.841    -2.242408    1.826734
              |
  campaignint |   1.445783   .6340778     2.28   0.023     .2021621    2.689404
       polint |   .9357577   .5748141     1.63   0.104     -.191629    2.063144
      polknow |   .0427671   .5592423     0.08   0.939    -1.054079    1.139613
       gender |   -.155069   .2469242    -0.63   0.530     -.639363     .329225
        White |  -.5171923   .2417324    -2.14   0.033    -.9913036   -.0430811
       latinx |  -1.197889   .4510212    -2.66   0.008     -2.08248   -.3132984
         ageN |  -.0819729   .8496745    -0.10   0.923    -1.748445    1.584499
    education |   1.037908   .5068216     2.05   0.041     .0438751     2.03194
income_norm01 |   -.001155   .4775441    -0.00   0.998    -.9377655    .9354556
        _cons |  -4.885737   1.087115    -4.49   0.000    -7.017902   -2.753572
-------------------------------------------------------------------------------

. outreg2 using tabDemhope12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemhope12.doc
dir : seeout

. svy: logit button c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,581
Number of PSUs   = 1,778                           Population size = 2,295.671
                                                   Design df       =     1,768
                                                   F(12, 1757)     =      8.81
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.151578   .4996229     2.30   0.021     .1716644    2.131492
        1.web |   .3366906   .4200057     0.80   0.423    -.4870694    1.160451
              |
web#c.hopeful |
           1  |   .0735758   .5948812     0.12   0.902    -1.093169     1.24032
              |
  campaignint |   1.376699    .337428     4.08   0.000     .7148992    2.038499
       polint |   .5052783   .3393219     1.49   0.137    -.1602359    1.170793
      polknow |  -.1470205    .358804    -0.41   0.682    -.8507452    .5567041
       gender |  -.0776539   .1612867    -0.48   0.630    -.3939866    .2386788
        White |  -.7516835    .169415    -4.44   0.000    -1.083958   -.4194087
       latinx |  -.7289724   .2293294    -3.18   0.002    -1.178758   -.2791871
         ageN |    .365506   .4908528     0.74   0.457    -.5972069    1.328219
    education |  -.5527573   .2968303    -1.86   0.063    -1.134933     .029418
income_norm01 |   .3736191   .2848038     1.31   0.190    -.1849686    .9322067
        _cons |  -3.160929   .5482638    -5.77   0.000    -4.236243   -2.085616
-------------------------------------------------------------------------------

. outreg2 using tabDemhope12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemhope12.doc
dir : seeout

. svy: logit volunteer c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if p
> artyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,581
Number of PSUs   = 1,778                           Population size = 2,295.671
                                                   Design df       =     1,768
                                                   F(12, 1757)     =      5.90
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   2.641767   .8736331     3.02   0.003     .9283048     4.35523
        1.web |   1.378612   .8081909     1.71   0.088    -.2064983    2.963722
              |
web#c.hopeful |
           1  |  -1.827246   1.073705    -1.70   0.089    -3.933111    .2786188
              |
  campaignint |   1.499913   .5529928     2.71   0.007     .4153249    2.584502
       polint |   1.279412   .7112452     1.80   0.072    -.1155581    2.674382
      polknow |  -.0411177   .6740414    -0.06   0.951     -1.36312    1.280884
       gender |  -.1434584   .2873695    -0.50   0.618    -.7070781    .4201613
        White |  -.3973204   .3098164    -1.28   0.200    -1.004965    .2103246
       latinx |  -.4503591   .4803372    -0.94   0.349    -1.392448    .4917295
         ageN |   1.787044   .9257218     1.93   0.054    -.0285805    3.602668
    education |   .7882865   .4846656     1.63   0.104    -.1622913    1.738864
income_norm01 |   .2679158   .4454599     0.60   0.548    -.6057676    1.141599
        _cons |  -7.940318   1.376127    -5.77   0.000    -10.63932   -5.241311
-------------------------------------------------------------------------------

. outreg2 using tabDemhope12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemhope12.doc
dir : seeout

. svy: logit donation c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if pa
> rtyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,579
Number of PSUs   = 1,778                           Population size = 2,294.911
                                                   Design df       =     1,768
                                                   F(12, 1757)     =     11.62
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   2.032446    .730778     2.78   0.005     .5991657    3.465725
        1.web |   .5653841   .6154953     0.92   0.358     -.641791    1.772559
              |
web#c.hopeful |
           1  |  -.2223099   .8346169    -0.27   0.790     -1.85925     1.41463
              |
  campaignint |   1.436407   .3643125     3.94   0.000     .7218781    2.150935
       polint |   .9542465   .4321063     2.21   0.027     .1067535    1.801739
      polknow |   .5772959   .5068819     1.14   0.255    -.4168549    1.571447
       gender |  -.1268133   .1900951    -0.67   0.505    -.4996481    .2460214
        White |  -.5589616   .2068943    -2.70   0.007    -.9647448   -.1531784
       latinx |  -.4710042   .3067484    -1.54   0.125    -1.072632    .1306236
         ageN |   3.048115    .611295     4.99   0.000     1.849178    4.247052
    education |   1.124728   .3473225     3.24   0.001     .4435218    1.805934
income_norm01 |   1.625833   .3518288     4.62   0.000     .9357889    2.315877
        _cons |  -8.057492   .8510527    -9.47   0.000    -9.726667   -6.388316
-------------------------------------------------------------------------------

. outreg2 using tabDemhope12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemhope12.doc
dir : seeout

. 
. *********************

. ** Table B10 -- 2012 **

. *********************

. 
. svy: logit persuade c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if pa
> rtyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
                                                   Design df       =     1,378
                                                   F(12, 1367)     =     14.79
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.077136   .3990042     2.70   0.007     .2944142    1.859857
        1.web |   -.189227   .2779329    -0.68   0.496    -.7344444    .3559904
              |
web#c.hopeful |
           1  |   .3246405   .4730419     0.69   0.493    -.6033195    1.252601
              |
  campaignint |   1.267798   .2815654     4.50   0.000     .7154546    1.820141
       polint |   .9040886   .3580323     2.53   0.012     .2017412    1.606436
      polknow |   .6463024   .3708085     1.74   0.082    -.0811079    1.373713
       gender |  -.1062252   .1395978    -0.76   0.447    -.3800723     .167622
        White |    .590599   .2555122     2.31   0.021     .0893639    1.091834
       latinx |   .6827616   .3707513     1.84   0.066    -.0445364     1.41006
         ageN |   .0152418   .4758337     0.03   0.974    -.9181951    .9486786
    education |   -.010632   .2679348    -0.04   0.968    -.5362362    .5149721
income_norm01 |   .3000846   .2720119     1.10   0.270    -.2335177    .8336869
        _cons |  -3.132272   .4509849    -6.95   0.000    -4.016964   -2.247581
-------------------------------------------------------------------------------

. outreg2 using tabRephope12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRephope12.doc
dir : seeout

. svy: logit rally c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if party
> id_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      6.92
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.452342   1.032148     1.41   0.160    -.5724093    3.477093
        1.web |   .8722277    .745673     1.17   0.242    -.5905494    2.335005
              |
web#c.hopeful |
           1  |  -1.366687   1.123163    -1.22   0.224    -3.569982    .8366074
              |
  campaignint |   .9816364   .8123048     1.21   0.227    -.6118513    2.575124
       polint |   1.974247   .8962066     2.20   0.028     .2161704    3.732324
      polknow |   .9511691   .9015172     1.06   0.292    -.8173254    2.719664
       gender |   .3332092   .2704702     1.23   0.218    -.1973688    .8637871
        White |  -.2903336   .4286546    -0.68   0.498     -1.13122    .5505526
       latinx |   .1749789   .7047025     0.25   0.804    -1.207427    1.557385
         ageN |    .349152   1.097933     0.32   0.751    -1.804648    2.502952
    education |   .8697571     .47066     1.85   0.065    -.0535305    1.793045
income_norm01 |   -.747213   .4997449    -1.50   0.135    -1.727556    .2331301
        _cons |   -6.77285   .9945013    -6.81   0.000    -8.723751    -4.82195
-------------------------------------------------------------------------------

. outreg2 using tabRephope12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRephope12.doc
dir : seeout

. svy: logit button c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      7.02
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   2.090116   .6826642     3.06   0.002     .7509429     3.42929
        1.web |   1.183822    .508496     2.33   0.020     .1863123    2.181332
              |
web#c.hopeful |
           1  |  -1.827522    .780256    -2.34   0.019     -3.35814   -.2969041
              |
  campaignint |   1.369632   .4071979     3.36   0.001     .5708369    2.168426
       polint |   1.868025   .4972684     3.76   0.000       .89254     2.84351
      polknow |  -.2886761   .5522453    -0.52   0.601    -1.372009    .7946562
       gender |   .1761754   .1829921     0.96   0.336    -.1827978    .5351486
        White |  -.0550565   .3143555    -0.18   0.861    -.6717237    .5616107
       latinx |  -.1182606   .5364665    -0.22   0.826     -1.17064    .9341188
         ageN |  -.3084147   .6297833    -0.49   0.624    -1.543852     .927023
    education |   -.452007   .3669832    -1.23   0.218    -1.171913    .2678991
income_norm01 |  -.6691551   .3393868    -1.97   0.049    -1.334926   -.0033845
        _cons |  -4.542773   .8142368    -5.58   0.000    -6.140051   -2.945495
-------------------------------------------------------------------------------

. outreg2 using tabRephope12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRephope12.doc
dir : seeout

. svy: logit volunteer c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if p
> artyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      4.69
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   2.164182   .8660481     2.50   0.013      .465267    3.863098
        1.web |   1.426062   .7712893     1.85   0.065    -.0869667     2.93909
              |
web#c.hopeful |
           1  |  -1.926334   1.086311    -1.77   0.076    -4.057336    .2046677
              |
  campaignint |   .9298247   .7378042     1.26   0.208    -.5175162    2.377166
       polint |    2.70171    .944538     2.86   0.004     .8488216    4.554598
      polknow |  -.0420047   1.270959    -0.03   0.974    -2.535229    2.451219
       gender |  -.0677544    .406433    -0.17   0.868    -.8650488      .72954
        White |   -.076756   .5472967    -0.14   0.888    -1.150381    .9968688
       latinx |  -.5984203    .742327    -0.81   0.420    -2.054633     .857793
         ageN |  -.2093345   1.748829    -0.12   0.905     -3.63999    3.221321
    education |   .2876302   .5406202     0.53   0.595    -.7728973    1.348158
income_norm01 |  -2.144826   .6047874    -3.55   0.000    -3.331229   -.9584219
        _cons |    -6.6496   1.368853    -4.86   0.000    -9.334861   -3.964339
-------------------------------------------------------------------------------

. outreg2 using tabRephope12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRephope12.doc
dir : seeout

. svy: logit donation c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if pa
> rtyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,705
Number of PSUs   = 1,388                           Population size = 1,973.483
                                                   Design df       =     1,378
                                                   F(12, 1367)     =      8.70
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.369575   .6160342     2.22   0.026     .1611085    2.578041
        1.web |   .3831643   .5115728     0.75   0.454    -.6203814     1.38671
              |
web#c.hopeful |
           1  |  -.3493733   .7478229    -0.47   0.640    -1.816368    1.117621
              |
  campaignint |    1.28202   .5279637     2.43   0.015     .2463202    2.317719
       polint |    2.39997   .5529774     4.34   0.000     1.315201    3.484738
      polknow |   1.800588   .6681534     2.69   0.007     .4898805    3.111296
       gender |   .2180344   .2071797     1.05   0.293    -.1883873     .624456
        White |  -.5784066   .3708864    -1.56   0.119     -1.30597    .1491565
       latinx |  -.5084925   .5194629    -0.98   0.328    -1.527516    .5105312
         ageN |   1.082382   .7720348     1.40   0.161     -.432109    2.596872
    education |   .5041413   .3661907     1.38   0.169    -.2142102    1.222493
income_norm01 |   .6098255   .4103913     1.49   0.138    -.1952337    1.414885
        _cons |  -7.840065   .8835575    -8.87   0.000    -9.573329   -6.106802
-------------------------------------------------------------------------------

. outreg2 using tabRephope12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRephope12.doc
dir : seeout

. 
. *********************

. ** Table B11 -- 2012 **

. *********************

. 
. svy: logit persuade c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,576
Number of PSUs   = 1,778                           Population size = 2,287.619
                                                   Design df       =     1,768
                                                   F(12, 1757)     =     13.79
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .3107015   .3275326     0.95   0.343    -.3316903    .9530933
        1.web |  -.0192214   .2742144    -0.07   0.944    -.5570399    .5185971
              |
  web#c.proud |
           1  |   .2026434   .3829945     0.53   0.597    -.5485262    .9538131
              |
  campaignint |   1.544469   .2388548     6.47   0.000     1.076001    2.012936
       polint |   1.326281   .2878043     4.61   0.000     .7618087    1.890754
      polknow |  -.2436974    .317259    -0.77   0.443    -.8659396    .3785448
       gender |  -.0682009   .1242182    -0.55   0.583     -.311831    .1754291
        White |  -.1847576   .1518453    -1.22   0.224    -.4825728    .1130575
       latinx |   -.300267   .2032055    -1.48   0.140    -.6988152    .0982812
         ageN |   .3280736   .3918337     0.84   0.403    -.4404325     1.09658
    education |  -.3157471   .2374105    -1.33   0.184    -.7813819    .1498877
income_norm01 |   .4056887   .2390172     1.70   0.090    -.0630974    .8744748
        _cons |  -2.387569   .3574824    -6.68   0.000    -3.088701   -1.686436
-------------------------------------------------------------------------------

. outreg2 using tabDempride12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDempride12.doc
dir : seeout

. svy: logit rally c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                            Number of obs   =    2,576
Number of PSUs   = 1,778                            Population size = 2,292.23
                                                    Design df       =    1,768
                                                    F(12, 1757)     =     5.35
                                                    Prob > F        =   0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |  -.1375704   .5967503    -0.23   0.818    -1.307981     1.03284
        1.web |    -.62519   .5523065    -1.13   0.258    -1.708432    .4580525
              |
  web#c.proud |
           1  |   .9185847   .7015039     1.31   0.191    -.4572795    2.294449
              |
  campaignint |   1.502269   .6276178     2.39   0.017     .2713177     2.73322
       polint |   .9371956   .5800637     1.62   0.106    -.2004872    2.074878
      polknow |   .0273626   .5547066     0.05   0.961    -1.060587    1.115312
       gender |  -.1537464   .2490327    -0.62   0.537     -.642176    .3346832
        White |  -.5304242   .2399574    -2.21   0.027    -1.001054   -.0597942
       latinx |  -1.206655   .4564816    -2.64   0.008    -2.101955   -.3113547
         ageN |  -.0805989   .8449311    -0.10   0.924    -1.737768     1.57657
    education |   1.068506   .4989507     2.14   0.032     .0899111    2.047102
income_norm01 |  -.0467349   .4754537    -0.10   0.922    -.9792453    .8857756
        _cons |  -4.258855   1.008515    -4.22   0.000    -6.236862   -2.280847
-------------------------------------------------------------------------------

. outreg2 using tabDempride12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDempride12.doc
dir : seeout

. svy: logit button c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyi
> d_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,577
Number of PSUs   = 1,778                           Population size = 2,292.532
                                                   Design df       =     1,768
                                                   F(12, 1757)     =      9.78
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.457926   .5233162     2.79   0.005     .4315427     2.48431
        1.web |   .3795068    .451833     0.84   0.401    -.5066762     1.26569
              |
  web#c.proud |
           1  |   .0539436   .5855393     0.09   0.927    -1.094478    1.202366
              |
  campaignint |   1.274746   .3412959     3.74   0.000     .6053597    1.944131
       polint |   .4603494   .3528711     1.30   0.192    -.2317392    1.152438
      polknow |  -.1019925   .3578308    -0.29   0.776    -.8038085    .5998235
       gender |  -.0760572   .1619117    -0.47   0.639    -.3936158    .2415014
        White |  -.6188097   .1681529    -3.68   0.000     -.948609   -.2890103
       latinx |   -.608421   .2314781    -2.63   0.009    -1.062421   -.1544213
         ageN |   .2592877   .5010351     0.52   0.605    -.7233958    1.241971
    education |  -.5316174   .2980363    -1.78   0.075    -1.116158    .0529232
income_norm01 |   .3326818   .2875484     1.16   0.247    -.2312887    .8966523
        _cons |   -3.34278   .5904098    -5.66   0.000    -4.500755   -2.184806
-------------------------------------------------------------------------------

. outreg2 using tabDempride12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDempride12.doc
dir : seeout

. svy: logit volunteer c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if par
> tyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,577
Number of PSUs   = 1,778                           Population size = 2,292.532
                                                   Design df       =     1,768
                                                   F(12, 1757)     =      5.29
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |      2.423   .9238438     2.62   0.009     .6110594    4.234941
        1.web |   1.027549   .8090373     1.27   0.204    -.5592208     2.61432
              |
  web#c.proud |
           1  |  -1.243482   1.016667    -1.22   0.221    -3.237478    .7505145
              |
  campaignint |   1.388416   .5481597     2.53   0.011     .3133064    2.463525
       polint |   1.212651   .7092158     1.71   0.087    -.1783392     2.60364
      polknow |   .0223124   .6766205     0.03   0.974    -1.304748    1.349373
       gender |  -.1615218   .2912547    -0.55   0.579    -.7327615     .409718
        White |  -.3449849   .2960685    -1.17   0.244     -.925666    .2356961
       latinx |  -.3828592   .4778551    -0.80   0.423     -1.32008    .5543611
         ageN |   1.740943   .9401213     1.85   0.064    -.1029237    3.584809
    education |   .8431962   .4846309     1.74   0.082    -.1073137    1.793706
income_norm01 |   .2077191   .4451562     0.47   0.641    -.6653687    1.080807
        _cons |  -7.743419   1.417488    -5.46   0.000    -10.52355    -4.96329
-------------------------------------------------------------------------------

. outreg2 using tabDempride12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDempride12.doc
dir : seeout

. svy: logit donation c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     2,575
Number of PSUs   = 1,778                           Population size = 2,291.772
                                                   Design df       =     1,768
                                                   F(12, 1757)     =     12.54
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .6873078   .5699299     1.21   0.228    -.4304995    1.805115
        1.web |   -.023855   .4769234    -0.05   0.960     -.959248    .9115381
              |
  web#c.proud |
           1  |   .6393547   .6310294     1.01   0.311    -.5982876    1.876997
              |
  campaignint |   1.532683   .3714496     4.13   0.000     .8041561    2.261209
       polint |   .9761744   .4320923     2.26   0.024     .1287088     1.82364
      polknow |   .5392914   .5038959     1.07   0.285     -.449003    1.527586
       gender |  -.1164197   .1923961    -0.61   0.545    -.4937675    .2609282
        White |  -.5672932   .2113075    -2.68   0.007    -.9817321   -.1528543
       latinx |  -.4355893   .3149479    -1.38   0.167    -1.053299    .1821202
         ageN |   2.976433   .6101052     4.88   0.000     1.779829    4.173036
    education |   1.121319   .3420142     3.28   0.001     .4505241    1.792114
income_norm01 |   1.503712    .354138     4.25   0.000     .8091389    2.198285
        _cons |   -7.14926   .7697912    -9.29   0.000    -8.659057   -5.639463
-------------------------------------------------------------------------------

. outreg2 using tabDempride12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDempride12.doc
dir : seeout

. 
. *********************

. ** Table B12 -- 2012 **

. *********************

. 
. svy: logit persuade c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
                                                   Design df       =     1,379
                                                   F(12, 1368)     =     14.84
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.374513   .4029781     3.41   0.001     .5839963    2.165029
        1.web |  -.0785034   .2273487    -0.35   0.730      -.52449    .3674833
              |
  web#c.proud |
           1  |  -.0370689   .4616971    -0.08   0.936    -.9427735    .8686357
              |
  campaignint |   1.234919     .28126     4.39   0.000     .6831749    1.786663
       polint |   .9211665   .3573222     2.58   0.010     .2202126     1.62212
      polknow |   .6841164    .370529     1.85   0.065    -.0427451    1.410978
       gender |   -.172293    .139406    -1.24   0.217    -.4457639    .1011779
        White |   .6127104   .2631747     2.33   0.020     .0964444    1.128976
       latinx |   .7349745   .3808917     1.93   0.054    -.0122153    1.482164
         ageN |   -.048065   .4839301    -0.10   0.921    -.9973839    .9012539
    education |   .0312533    .266735     0.12   0.907    -.4919968    .5545035
income_norm01 |   .3515714   .2753107     1.28   0.202    -.1885018    .8916445
        _cons |  -3.077518   .4369496    -7.04   0.000    -3.934676    -2.22036
-------------------------------------------------------------------------------

. outreg2 using tabReppride12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabReppride12.doc
dir : seeout

. svy: logit rally c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid
> _2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      6.55
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .6248918   .7913435     0.79   0.430    -.9274755    2.177259
        1.web |   .1144454   .5052366     0.23   0.821      -.87667    1.105561
              |
  web#c.proud |
           1  |   .2573813    .883072     0.29   0.771    -1.474929    1.989691
              |
  campaignint |   1.002578   .8613417     1.16   0.245    -.6871042    2.692259
       polint |   1.805606   .9861658     1.83   0.067    -.1289418    3.740153
      polknow |   .9864347   .9458566     1.04   0.297    -.8690387    2.841908
       gender |   .3577564   .2712555     1.32   0.187    -.1743616    .8898743
        White |  -.3949455   .4199271    -0.94   0.347    -1.218711    .4288195
       latinx |   .0117897   .7306926     0.02   0.987      -1.4216    1.445179
         ageN |  -.2339589   1.147724    -0.20   0.839    -2.485432    2.017514
    education |   .7060915   .4802526     1.47   0.142    -.2360131    1.648196
income_norm01 |  -.6240241   .5380951    -1.16   0.246    -1.679598    .4315495
        _cons |  -5.877788   .9271692    -6.34   0.000    -7.696603   -4.058974
-------------------------------------------------------------------------------

. outreg2 using tabReppride12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabReppride12.doc
dir : seeout

. svy: logit button c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyi
> d_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      7.42
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.477868   .5484434     2.69   0.007     .4019941    2.553741
        1.web |   .5508084   .3465669     1.59   0.112    -.1290469    1.230664
              |
  web#c.proud |
           1  |  -.8900297   .6167689    -1.44   0.149    -2.099936     .319877
              |
  campaignint |   1.413899   .4135121     3.42   0.001     .6027184     2.22508
       polint |   1.716884    .509019     3.37   0.001     .7183489     2.71542
      polknow |   -.274043   .5574198    -0.49   0.623    -1.367526    .8194396
       gender |   .1545703   .1915283     0.81   0.420    -.2211481    .5302887
        White |  -.1400815   .3138362    -0.45   0.655    -.7557296    .4755665
       latinx |    -.22335   .5310361    -0.42   0.674    -1.265076    .8183759
         ageN |  -.5892879   .6468149    -0.91   0.362    -1.858135    .6795597
    education |  -.6280902   .3688825    -1.70   0.089    -1.351722    .0955414
income_norm01 |  -.5664775   .3508388    -1.61   0.107    -1.254713     .121758
        _cons |  -3.697182   .6295917    -5.87   0.000    -4.932243   -2.462121
-------------------------------------------------------------------------------

. outreg2 using tabReppride12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabReppride12.doc
dir : seeout

. svy: logit volunteer c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if par
> tyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      5.73
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   7.587754   2.879478     2.64   0.009     1.939122    13.23639
        1.web |    5.22585   2.321585     2.25   0.025     .6716298    9.780069
              |
  web#c.proud |
           1  |  -6.236696   2.934677    -2.13   0.034    -11.99361   -.4797819
              |
  campaignint |   1.022813    .858422     1.19   0.234    -.6611413    2.706767
       polint |   2.367702   1.052717     2.25   0.025     .3026024    4.432801
      polknow |   .2045996   1.184212     0.17   0.863    -2.118453    2.527652
       gender |   .1019049   .3755146     0.27   0.786    -.6347368    .8385466
        White |   -.232624   .5923279    -0.39   0.695    -1.394585    .9293373
       latinx |  -.9216595   .8227725    -1.12   0.263    -2.535681    .6923615
         ageN |  -1.866144   1.691653    -1.10   0.270    -5.184636    1.452347
    education |   .1716704     .55173     0.31   0.756    -.9106505    1.253991
income_norm01 |  -1.974962   .6595057    -2.99   0.003    -3.268705   -.6812195
        _cons |  -10.08137   2.823301    -3.57   0.000     -15.6198   -4.542938
-------------------------------------------------------------------------------

. outreg2 using tabReppride12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabReppride12.doc
dir : seeout

. svy: logit donation c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if part
> yid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =    10                           Number of obs   =     1,697
Number of PSUs   = 1,389                           Population size = 1,958.877
                                                   Design df       =     1,379
                                                   F(12, 1368)     =      9.03
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     donation | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.074765   .7211978     1.49   0.136    -.3399981    2.489529
        1.web |   .1495541   .3992137     0.37   0.708    -.6335777    .9326859
              |
  web#c.proud |
           1  |   .0276719   .7833212     0.04   0.972    -1.508958    1.564302
              |
  campaignint |   1.215953   .5600815     2.17   0.030     .1172493    2.314657
       polint |   2.266979   .5593548     4.05   0.000       1.1697    3.364257
      polknow |   1.721819    .655727     2.63   0.009     .4354888    3.008149
       gender |   .1750116   .2113604     0.83   0.408    -.2396111    .5896342
        White |  -.5900464   .3651453    -1.62   0.106    -1.306347    .1262539
       latinx |  -.5403454   .5146406    -1.05   0.294    -1.549908    .4692177
         ageN |   1.025483   .7762632     1.32   0.187    -.4973013    2.548268
    education |   .4615656   .3725064     1.24   0.216    -.2691749    1.192306
income_norm01 |   .6803721   .4177539     1.63   0.104    -.1391298    1.499874
        _cons |   -7.27613    .789305    -9.22   0.000    -8.824498   -5.727762
-------------------------------------------------------------------------------

. outreg2 using tabReppride12.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabReppride12.doc
dir : seeout

. 
. *********************

. ** Figure C1 -- 2012 **

. *********************

. 
. svy: reg angrycandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat=
> =0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,756
Number of PSUs   = 1,904                           Population size = 2,454.208
                                                   Design df       =     1,894
                                                   F(10, 1885)     =     14.86
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.0820

-------------------------------------------------------------------------------
              |             Linearized
 angrycandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0346544   .0111475     3.11   0.002     .0127918    .0565171
  campaignint |   .1085872   .0179623     6.05   0.000     .0733593    .1438151
       polint |   .0766583   .0244271     3.14   0.002     .0287515    .1245652
      polknow |  -.0031488   .0276677    -0.11   0.909    -.0574111    .0511135
       gender |   .0192883   .0104536     1.85   0.065    -.0012134    .0397901
        White |   .0207565   .0123492     1.68   0.093     -.003463     .044976
       latinx |  -.0078433   .0170024    -0.46   0.645    -.0411886    .0255021
         ageN |  -.0675058   .0303014    -2.23   0.026    -.1269333   -.0080782
    education |   .0147886   .0202071     0.73   0.464     -.024842    .0544192
income_norm01 |   .0007424   .0203993     0.04   0.971    -.0392652    .0407499
        _cons |   .0977606   .0253401     3.86   0.000     .0480632    .1474579
-------------------------------------------------------------------------------

. estimates store angryDeavg

. svy: reg angrycandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat=
> =1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,830
Number of PSUs   = 1,488                           Population size = 2,115.152
                                                   Design df       =     1,478
                                                   F(10, 1469)     =     19.06
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1161

-------------------------------------------------------------------------------
              |             Linearized
 angrycandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0269577   .0112674     2.39   0.017     .0048559    .0490596
  campaignint |   .0748364     .01874     3.99   0.000     .0380766    .1115963
       polint |   .1304583   .0249291     5.23   0.000      .081558    .1793585
      polknow |  -.0153131   .0279985    -0.55   0.585    -.0702342    .0396079
       gender |   .0006971   .0094691     0.07   0.941    -.0178773    .0192715
        White |   .0308563   .0224072     1.38   0.169    -.0130969    .0748095
       latinx |   .0066511   .0288394     0.23   0.818    -.0499195    .0632217
         ageN |   .0320039   .0329139     0.97   0.331     -.032559    .0965667
    education |  -.0305114   .0198699    -1.54   0.125    -.0694876    .0084648
income_norm01 |  -.0102811   .0171682    -0.60   0.549    -.0439578    .0233955
        _cons |    .098534   .0314246     3.14   0.002     .0368925    .1601756
-------------------------------------------------------------------------------

. estimates store angryReavg

. coefplot (angryDeavg) || (angryReavg), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN educatio
> n income_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Anger toward Presidential Candidat
> es") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(angeravg12)

. 
. svy: reg afraidcandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat
> ==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,759
Number of PSUs   = 1,904                           Population size = 2,450.882
                                                   Design df       =     1,894
                                                   F(10, 1885)     =     11.12
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.0597

-------------------------------------------------------------------------------
              |             Linearized
afraidcandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0314203   .0106488     2.95   0.003     .0105357    .0523049
  campaignint |    .082284   .0176389     4.66   0.000     .0476902    .1168779
       polint |   .0619703   .0241607     2.56   0.010      .014586    .1093547
      polknow |  -.0363181   .0255603    -1.42   0.156    -.0864473    .0138112
       gender |   .0051171   .0099389     0.51   0.607    -.0143753    .0246094
        White |   .0181496   .0116835     1.55   0.120    -.0047642    .0410635
       latinx |  -.0374137   .0146255    -2.56   0.011    -.0660975   -.0087298
         ageN |  -.0157287   .0312258    -0.50   0.615    -.0769692    .0455119
    education |  -.0022131   .0187614    -0.12   0.906    -.0390082     .034582
income_norm01 |   .0105329   .0201362     0.52   0.601    -.0289585    .0500243
        _cons |   .0757985   .0238535     3.18   0.002     .0290167    .1225804
-------------------------------------------------------------------------------

. estimates store afraidDeavg

. svy: reg afraidcandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat
> ==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,836
Number of PSUs   = 1,490                           Population size = 2,123.453
                                                   Design df       =     1,480
                                                   F(10, 1471)     =     19.61
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1118

-------------------------------------------------------------------------------
              |             Linearized
afraidcandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0690418   .0111618     6.19   0.000     .0471472    .0909364
  campaignint |    .073975   .0199044     3.72   0.000     .0349312    .1130188
       polint |   .0913567   .0262636     3.48   0.001     .0398389    .1428744
      polknow |   .0231738   .0282959     0.82   0.413    -.0323306    .0786783
       gender |   .0238804   .0104163     2.29   0.022     .0034481    .0443127
        White |   .0170782    .023025     0.74   0.458    -.0280868    .0622433
       latinx |  -.0097283   .0301936    -0.32   0.747    -.0689551    .0494986
         ageN |   .0515873   .0327948     1.57   0.116    -.0127419    .1159164
    education |  -.0311405   .0212092    -1.47   0.142    -.0727438    .0104629
income_norm01 |  -.0187683    .019139    -0.98   0.327    -.0563107    .0187741
        _cons |   .0088431    .030556     0.29   0.772    -.0510946    .0687809
-------------------------------------------------------------------------------

. estimates store afraidReavg

. coefplot (afraidDeavg) || (afraidReavg), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN educat
> ion income_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Fear toward Presidential Candida
> tes") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(fearavg12)

. 
. svy: reg hopecandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==
> 0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,756
Number of PSUs   = 1,903                           Population size = 2,452.851
                                                   Design df       =     1,893
                                                   F(10, 1884)     =     24.24
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1337

-------------------------------------------------------------------------------
              |             Linearized
  hopecandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |  -.0368655   .0093959    -3.92   0.000    -.0552928   -.0184382
  campaignint |   .0821247   .0143315     5.73   0.000     .0540176    .1102318
       polint |   .0889239   .0176975     5.02   0.000     .0542154    .1236325
      polknow |  -.0138114   .0218602    -0.63   0.528    -.0566841    .0290612
       gender |    .024246    .008388     2.89   0.004     .0077952    .0406968
        White |  -.0577144   .0102486    -5.63   0.000    -.0778142   -.0376146
       latinx |  -.0262604   .0131469    -2.00   0.046    -.0520443   -.0004765
         ageN |   .0211583   .0235625     0.90   0.369    -.0250528    .0673694
    education |   .0094434   .0157983     0.60   0.550    -.0215404    .0404273
income_norm01 |   .0000777   .0172817     0.00   0.996    -.0338155    .0339709
        _cons |    .221044   .0217192    10.18   0.000     .1784479      .26364
-------------------------------------------------------------------------------

. estimates store hopefulDeavg

. svy: reg hopecandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==
> 1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,828
Number of PSUs   = 1,489                           Population size = 2,111.458
                                                   Design df       =     1,479
                                                   F(10, 1470)     =     16.77
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1303

-------------------------------------------------------------------------------
              |             Linearized
  hopecandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |  -.0353153   .0107673    -3.28   0.001    -.0564361   -.0141946
  campaignint |   .0891332    .017816     5.00   0.000     .0541858    .1240806
       polint |   .0931237   .0238336     3.91   0.000     .0463724     .139875
      polknow |    .003294   .0259858     0.13   0.899    -.0476789    .0542669
       gender |    .050285   .0093442     5.38   0.000     .0319558    .0686142
        White |  -.0072417   .0234046    -0.31   0.757    -.0531514    .0386679
       latinx |   .0122343   .0306323     0.40   0.690    -.0478531    .0723217
         ageN |   .0096506   .0297081     0.32   0.745    -.0486238    .0679251
    education |  -.0114394   .0188863    -0.61   0.545    -.0484861    .0256074
income_norm01 |   .0192876   .0173219     1.11   0.266    -.0146905    .0532657
        _cons |   .1544767   .0354598     4.36   0.000     .0849198    .2240336
-------------------------------------------------------------------------------

. estimates store hopefulReavg

. coefplot (hopefulDeavg) || (hopefulReavg), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN educ
> ation income_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Hope toward Presidential Candi
> dates") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(hopeavg12)

. 
. svy: reg pridecandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat=
> =0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     2,744
Number of PSUs   = 1,902                           Population size = 2,444.508
                                                   Design df       =     1,892
                                                   F(10, 1883)     =     36.09
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1786

-------------------------------------------------------------------------------
              |             Linearized
 pridecandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |  -.0319061   .0111243    -2.87   0.004    -.0537233   -.0100889
  campaignint |   .1202047   .0162772     7.38   0.000     .0882816    .1521278
       polint |   .0851723   .0210078     4.05   0.000     .0439713    .1263733
      polknow |  -.0272146   .0233103    -1.17   0.243    -.0729311     .018502
       gender |   .0222227   .0092143     2.41   0.016     .0041514     .040294
        White |   -.094388    .011246    -8.39   0.000    -.1164439   -.0723321
       latinx |  -.0683704   .0140115    -4.88   0.000      -.09585   -.0408908
         ageN |   .0542176   .0270522     2.00   0.045     .0011622    .1072729
    education |   .0026844   .0169708     0.16   0.874    -.0305992    .0359679
income_norm01 |   .0167679   .0179289     0.94   0.350    -.0183946    .0519305
        _cons |   .2018512   .0240926     8.38   0.000     .1546004     .249102
-------------------------------------------------------------------------------

. estimates store proudDeavg

. svy: reg pridecandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat=
> =1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =    10                           Number of obs   =     1,818
Number of PSUs   = 1,490                           Population size = 2,093.837
                                                   Design df       =     1,480
                                                   F(10, 1471)     =     21.72
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1488

-------------------------------------------------------------------------------
              |             Linearized
 pridecandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |  -.0151094   .0123269    -1.23   0.220    -.0392895    .0090707
  campaignint |   .1006791   .0206769     4.87   0.000     .0601199    .1412382
       polint |   .1181398   .0259948     4.54   0.000     .0671492    .1691303
      polknow |  -.0262241   .0308214    -0.85   0.395    -.0866825    .0342342
       gender |   .0719706   .0110624     6.51   0.000      .050271    .0936702
        White |  -.0031146   .0223343    -0.14   0.889    -.0469249    .0406957
       latinx |   .0269074   .0297087     0.91   0.365    -.0313682    .0851831
         ageN |   .0550338   .0332643     1.65   0.098    -.0102164    .1202841
    education |   .0116344   .0204635     0.57   0.570    -.0285061     .051775
income_norm01 |  -.0028072   .0188702    -0.15   0.882    -.0398224     .034208
        _cons |   .0361421   .0327341     1.10   0.270     -.028068    .1003522
-------------------------------------------------------------------------------

. estimates store proudReavg

. coefplot (proudDeavg) || (proudReavg), scheme(plottig) legend(off) drop(?cons ageN campaignint polint polknow educatio
> n income_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Pride toward Presidential Candidat
> es") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(prideavg12)

. 
. graph combine angeravg12 fearavg12 hopeavg12 prideavg12, scheme(plottig) xcommon ycommon col(2)

. log close
      name:  <unnamed>
       log:  /Users/marziaoceno/Dropbox/How Social Desirability Bias Impacts the Expression of Emotions.log
  log type:  text
 closed on:  10 May 2025, 14:43:52
------------------------------------------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------------------------------------------------------------------------------
      name:  <unnamed>
       log:  /Users/marziaoceno/Dropbox/How Social Desirability Bias Impacts the Expression of Emotions.log
  log type:  text
 opened on:  10 May 2025, 14:47:08

. *******************************ANES 2016*****************************************

. *********************

. ** Figure 2 -- 2016 **

. *********************

. 
. cibar anger [aweight=V160101], over(web partyid_2cat) graphopts(title("Anger") ytitle("Mean Anger toward Out-Party Cand.") scheme(plotplain) name(anger16, repl
> ace))

. cibar fear [aweight=V160101], over(web partyid_2cat) graphopts(title("Fear") ytitle("Mean Fear toward Out-Party Cand.")  scheme(plotplain) name(fear16, replace
> ))

. cibar hopeful [aweight=V160101], over(web partyid_2cat) graphopts(title("Hope") ytitle("Mean Hope toward In-Party Candid.")  scheme(plotplain) name(hope16, rep
> lace))

. cibar proud [aweight=V160101], over(web partyid_2cat) graphopts(title("Pride") ytitle("Mean Pride toward In-Party Candid.")  scheme(plotplain) name(pride16, re
> place))

. 
. graph combine anger16 fear16 hope16 pride16, scheme(plotplain) xcommon ycommon col(2)

. 
. 
. *********************

. ** Figure 4 -- 2016 **

. *********************

. 
. svyset [pweight=V160101]

Sampling weights: V160101
             VCE: linearized
     Single unit: missing
        Strata 1: <one>
 Sampling unit 1: <observations>
           FPC 1: <zero>

. 
. svy: reg anger web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                           Number of obs   =     1,791
Number of PSUs   = 1,791                           Population size = 1,829.147
                                                   Design df       =     1,790
                                                   F(10, 1781)     =     15.20
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1026

-------------------------------------------------------------------------------
              |             Linearized
        anger | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .1002454   .0198505     5.05   0.000     .0613129    .1391779
  campaignint |   .1180734   .0341675     3.46   0.001      .051061    .1850859
       polint |   .0673289   .0425247     1.58   0.114    -.0160744    .1507321
      polknow |   .0458501   .0345056     1.33   0.184    -.0218254    .1135256
       gender |   .0799393   .0179888     4.44   0.000      .044658    .1152206
        White |  -.1022219   .0212746    -4.80   0.000    -.1439475   -.0604963
       latinx |   .0364742   .0273761     1.33   0.183    -.0172183    .0901666
         ageN |  -.1145268   .0507102    -2.26   0.024    -.2139841   -.0150695
    education |   .1095012   .0597143     1.83   0.067    -.0076159    .2266183
income_norm01 |   .0908028   .0349848     2.60   0.010     .0221875     .159418
        _cons |   .4258539   .0505375     8.43   0.000     .3267352    .5249726
-------------------------------------------------------------------------------

. estimates store angryD

. svy: reg anger web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,576
Number of PSUs   = 1,576                          Population size = 1,538.0834
                                                  Design df       =      1,575
                                                  F(10, 1566)     =       7.78
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0751

-------------------------------------------------------------------------------
              |             Linearized
        anger | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0643238   .0211233     3.05   0.002      .022891    .1057566
  campaignint |   .0696518   .0359015     1.94   0.053    -.0007681    .1400716
       polint |   .1645164   .0481131     3.42   0.001     .0701439     .258889
      polknow |  -.0128353   .0357875    -0.36   0.720    -.0830314    .0573608
       gender |   .0222158   .0189673     1.17   0.242     -.014988    .0594197
        White |   .1401492   .0420321     3.33   0.001     .0577045    .2225939
       latinx |  -.0480113    .059857    -0.80   0.423    -.1654192    .0693965
         ageN |  -.1162497     .05657    -2.05   0.040    -.2272101   -.0052893
    education |  -.2037977   .0688592    -2.96   0.003    -.3388631   -.0687322
income_norm01 |   .0115735   .0372549     0.31   0.756    -.0615009    .0846478
        _cons |   .5221726   .0665867     7.84   0.000     .3915647    .6527805
-------------------------------------------------------------------------------

. estimates store angryR

. coefplot (angryD) || (angryR), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN education income_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "0"
>  .1 ".1" .2 ".2") xline(0) xtitle("Anger toward Out-Party Candidate") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(angerplot16)

. 
. svy: reg fear web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,788
Number of PSUs   = 1,788                          Population size = 1,828.4554
                                                  Design df       =      1,787
                                                  F(10, 1778)     =      11.50
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0841

-------------------------------------------------------------------------------
              |             Linearized
         fear | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .1172374   .0216659     5.41   0.000     .0747443    .1597305
  campaignint |   .1304352    .037995     3.43   0.001      .055916    .2049544
       polint |   .0157443   .0483002     0.33   0.744    -.0789865    .1104751
      polknow |   .0240438    .036862     0.65   0.514    -.0482534     .096341
       gender |   .0620864   .0199257     3.12   0.002     .0230063    .1011666
        White |  -.0917422    .022868    -4.01   0.000     -.136593   -.0468914
       latinx |  -.0014586   .0338517    -0.04   0.966    -.0678517    .0649345
         ageN |   .0701162   .0574481     1.22   0.222    -.0425564    .1827888
    education |   .1990571   .0640239     3.11   0.002     .0734875    .3246267
income_norm01 |   .0783163   .0376784     2.08   0.038     .0044178    .1522147
        _cons |   .2881705   .0551462     5.23   0.000     .1800127    .3963283
-------------------------------------------------------------------------------

. estimates store afraidD

. svy: reg fear web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,579
Number of PSUs   = 1,579                          Population size = 1,539.3775
                                                  Design df       =      1,578
                                                  F(10, 1569)     =       8.43
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0686

-------------------------------------------------------------------------------
              |             Linearized
         fear | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0603229   .0224571     2.69   0.007     .0162741    .1043718
  campaignint |   .0654347   .0377978     1.73   0.084    -.0087045    .1395739
       polint |   .1965942   .0478007     4.11   0.000     .1028345    .2903538
      polknow |  -.0242201    .037575    -0.64   0.519    -.0979224    .0494821
       gender |   .0281047   .0205994     1.36   0.173    -.0123004    .0685097
        White |   .1101438   .0448574     2.46   0.014     .0221574    .1981302
       latinx |   -.044474    .060102    -0.74   0.459    -.1623623    .0734142
         ageN |  -.0030536   .0621142    -0.05   0.961    -.1248887    .1187815
    education |  -.2422674   .0717565    -3.38   0.001    -.3830155   -.1015192
income_norm01 |  -.0357518     .03853    -0.93   0.354    -.1113271    .0398236
        _cons |   .4853919   .0724703     6.70   0.000     .3432438      .62754
-------------------------------------------------------------------------------

. estimates store afraidR

. coefplot (afraidD) || (afraidR), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN education income_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "
> 0" .1 ".1" .2 ".2") xline(0) xtitle("Fear toward Out-Party Candidate") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(fearplot16)

. 
. svy: reg hopeful web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                           Number of obs   =     1,790
Number of PSUs   = 1,790                           Population size = 1,829.837
                                                   Design df       =     1,789
                                                   F(10, 1780)     =     13.17
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1047

-------------------------------------------------------------------------------
              |             Linearized
      hopeful | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0151073   .0186831     0.81   0.419    -.0215357    .0517503
  campaignint |   .2071938   .0350935     5.90   0.000     .1383651    .2760224
       polint |   .0343754   .0445174     0.77   0.440    -.0529361     .121687
      polknow |   .0470729   .0328034     1.43   0.151    -.0172642    .1114099
       gender |   .0326876   .0184603     1.77   0.077    -.0035184    .0688937
        White |  -.1073595   .0219096    -4.90   0.000    -.1503305   -.0643885
       latinx |   .0025336   .0313089     0.08   0.936    -.0588723    .0639395
         ageN |    .179551   .0537775     3.34   0.001     .0740776    .2850243
    education |   .0222804   .0600718     0.37   0.711     -.095538    .1400987
income_norm01 |   .0130548   .0345412     0.38   0.706    -.0546906    .0808002
        _cons |   .2569812   .0512017     5.02   0.000     .1565598    .3574027
-------------------------------------------------------------------------------

. estimates store hopefulD

. svy: reg hopeful web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,578                          Population size = 1,538.8918
                                                  Design df       =      1,577
                                                  F(10, 1568)     =      21.17
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.1433

-------------------------------------------------------------------------------
              |             Linearized
      hopeful | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0636713   .0203234     3.13   0.002     .0238076     .103535
  campaignint |   .1956429   .0359124     5.45   0.000     .1252018     .266084
       polint |   .1507674   .0466255     3.23   0.001      .059313    .2422218
      polknow |  -.0837815   .0329935    -2.54   0.011    -.1484971   -.0190659
       gender |  -.0216152   .0187908    -1.15   0.250    -.0584727    .0152423
        White |   .0917623   .0381754     2.40   0.016     .0168824    .1666423
       latinx |  -.0362831   .0565441    -0.64   0.521    -.1471926    .0746264
         ageN |   .0650697   .0569422     1.14   0.253    -.0466207    .1767601
    education |  -.3512986   .0680686    -5.16   0.000    -.4848131   -.2177842
income_norm01 |  -.0601919   .0349536    -1.72   0.085    -.1287523    .0083686
        _cons |   .4331723   .0678643     6.38   0.000     .3000584    .5662861
-------------------------------------------------------------------------------

. estimates store hopefulR

. coefplot (hopefulD) || (hopefulR), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN education income_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0
>  "0" .1 ".1" .2 ".2") xline(0) xtitle("Hope toward In-Party Candidate") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(hopeplot16)

. 
. svy: reg proud web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,792
Number of PSUs   = 1,792                          Population size = 1,830.7972
                                                  Design df       =      1,791
                                                  F(10, 1782)     =      17.01
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.1345

-------------------------------------------------------------------------------
              |             Linearized
        proud | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   -.006734   .0192136    -0.35   0.726    -.0444174    .0309495
  campaignint |   .2195434   .0356995     6.15   0.000     .1495262    .2895605
       polint |   .0520017   .0460624     1.13   0.259      -.03834    .1423434
      polknow |   .0903791   .0334713     2.70   0.007     .0247321     .156026
       gender |   .0513276   .0183766     2.79   0.005     .0152857    .0873694
        White |  -.1440548   .0219135    -6.57   0.000    -.1870334   -.1010761
       latinx |  -.0125866   .0305611    -0.41   0.680    -.0725258    .0473526
         ageN |   .1754688   .0536617     3.27   0.001     .0702226     .280715
    education |    .003659   .0616692     0.06   0.953    -.1172922    .1246101
income_norm01 |  -.0259128   .0359619    -0.72   0.471    -.0964445     .044619
        _cons |   .2562704   .0514707     4.98   0.000     .1553214    .3572193
-------------------------------------------------------------------------------

. estimates store proudD

. svy: reg proud web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,578                          Population size = 1,537.8061
                                                  Design df       =      1,577
                                                  F(10, 1568)     =      19.33
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.1299

-------------------------------------------------------------------------------
              |             Linearized
        proud | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0818275   .0195595     4.18   0.000     .0434622    .1201928
  campaignint |   .1796924   .0355623     5.05   0.000     .1099379    .2494468
       polint |   .1294518   .0473448     2.73   0.006     .0365865    .2223172
      polknow |   -.114106   .0339039    -3.37   0.001    -.1806076   -.0476045
       gender |  -.0177048   .0184858    -0.96   0.338    -.0539642    .0185545
        White |   .0808625   .0365035     2.22   0.027      .009262    .1524629
       latinx |   -.041043   .0541088    -0.76   0.448    -.1471757    .0650898
         ageN |   .0731022   .0593749     1.23   0.218    -.0433599    .1895643
    education |    -.33945   .0692129    -4.90   0.000    -.4752089   -.2036911
income_norm01 |  -.0435116   .0352963    -1.23   0.218    -.1127441     .025721
        _cons |    .363234   .0648724     5.60   0.000     .2359889    .4904792
-------------------------------------------------------------------------------

. estimates store proudR

. coefplot (proudD) || (proudR), scheme(plottig) legend(off) drop(?cons ageN campaignint polint polknow education income_norm01) xlabel(-.2 "-.2" -.1 "-.1" 0 "0"
>  .1 ".1" .2 ".2") xline(0) xtitle("Pride toward In-Party Candidate") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(prideplot16)

. 
. graph combine angerplot16 fearplot16 hopeplot16 prideplot16, scheme(plottig) xcommon ycommon col(2)

. 
. 
. *********************

. ** Figure 6 -- 2016 **

. *********************

. 
. svyset [pweight=V160102]

Sampling weights: V160102
             VCE: linearized
     Single unit: missing
        Strata 1: <one>
 Sampling unit 1: <observations>
           FPC 1: <zero>

. 
. svy: logit persuade c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,599.9581
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       9.34
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .5343922   .3702504     1.44   0.149    -.1917663    1.260551
        1.web |  -.1002641   .3221981    -0.31   0.756    -.7321793    .5316512
              |
  web#c.anger |
           1  |   .1178166   .4431134     0.27   0.790    -.7512456    .9868787
              |
  campaignint |   1.354049   .2647129     5.12   0.000     .8348776    1.873221
       polint |    .592086   .3296837     1.80   0.073    -.0545106    1.238683
      polknow |   .7146296   .2570783     2.78   0.005     .2104313    1.218828
       gender |   .1022145   .1401692     0.73   0.466    -.1726942    .3771232
        White |    .277535   .1586649     1.75   0.080    -.0336487    .5887187
       latinx |   .0750466   .2339252     0.32   0.748    -.3837422    .5338354
         ageN |  -1.094857   .4150844    -2.64   0.008    -1.908947   -.2807677
    education |   .4571222   .4916381     0.93   0.353    -.5071097    1.421354
income_norm01 |   .0479174   .2593299     0.18   0.853    -.4606969    .5565316
        _cons |  -1.974852   .4282101    -4.61   0.000    -2.814685   -1.135019
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file1D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,791
Number of PSUs   = 1,828                          Population size = 1,599.9581
                                                  Subpop. no. obs =      1,566
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,827

Expression: Pr(persuade), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .1182428   .0802718     1.47   0.141    -.0391914     .275677
          2  |   .1438547   .0562398     2.56   0.011     .0335536    .2541558
------------------------------------------------------------------------------

. svy: logit rally c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,827
Number of PSUs   = 1,827                          Population size = 1,598.9989
                                                  Design df       =      1,826
                                                  F(12, 1815)     =       3.56
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |  -.7601046   .5765898    -1.32   0.188    -1.890949    .3707403
        1.web |  -.9586753   .5945461    -1.61   0.107    -2.124737    .2073865
              |
  web#c.anger |
           1  |   1.184198   .7657217     1.55   0.122    -.3175842    2.685981
              |
  campaignint |   1.503391   .5756642     2.61   0.009     .3743619    2.632421
       polint |    .925496   .7226669     1.28   0.200    -.4918446    2.342836
      polknow |    .078918   .4083663     0.19   0.847    -.7219962    .8798321
       gender |   .0435489   .2145492     0.20   0.839    -.3772387    .4643366
        White |   .1849436     .28522     0.65   0.517    -.3744482    .7443355
       latinx |  -.0191288   .4312576    -0.04   0.965    -.8649388    .8266812
         ageN |  -2.207567   .7116209    -3.10   0.002    -3.603243   -.8118905
    education |   .8136993   .8758531     0.93   0.353    -.9040798    2.531479
income_norm01 |  -.3962659   .4215209    -0.94   0.347     -1.22298    .4304479
        _cons |  -3.015866    .755731    -3.99   0.000    -4.498054   -1.533678
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file2D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,790
Number of PSUs   = 1,827                          Population size = 1,598.9989
                                                  Subpop. no. obs =      1,565
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,826

Expression: Pr(rally), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |  -.0636599   .0478132    -1.33   0.183    -.1574342    .0301144
          2  |   .0326885   .0397164     0.82   0.411    -.0452059    .1105828
------------------------------------------------------------------------------

. svy: logit button c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,599.9581
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       4.85
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .7033358   .6524924     1.08   0.281    -.5763736    1.983045
        1.web |   .0425708   .6147833     0.07   0.945    -1.163181    1.248323
              |
  web#c.anger |
           1  |   .3001683   .7748193     0.39   0.699    -1.219456    1.819793
              |
  campaignint |   1.398399   .3809456     3.67   0.000     .6512644    2.145534
       polint |   .8455602   .4922229     1.72   0.086    -.1198184    1.810939
      polknow |   .6251417   .3214947     1.94   0.052    -.0053942    1.255678
       gender |   .0388982   .1968609     0.20   0.843    -.3471979    .4249943
        White |   -.057111    .222117    -0.26   0.797    -.4927409    .3785189
       latinx |  -.4508434   .3436542    -1.31   0.190     -1.12484    .2231529
         ageN |  -1.411493    .581418    -2.43   0.015    -2.551807   -.2711796
    education |  -.6897887   .6352446    -1.09   0.278    -1.935671    .5560933
income_norm01 |  -.1364993   .3362699    -0.41   0.685    -.7960131    .5230146
        _cons |  -3.225653   .7357481    -4.38   0.000    -4.668649   -1.782657
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file3D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,791
Number of PSUs   = 1,828                          Population size = 1,599.9581
                                                  Subpop. no. obs =      1,566
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,827

Expression: Pr(button), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .0729371    .069126     1.06   0.292    -.0626372    .2085115
          2  |    .123711   .0532818     2.32   0.020     .0192114    .2282106
------------------------------------------------------------------------------

. svy: logit volunteer c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,599.9581
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       3.93
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.026981   1.136845     0.90   0.366    -1.202672    3.256633
        1.web |   .8148207   1.109301     0.73   0.463    -1.360811    2.990452
              |
  web#c.anger |
           1  |  -.8229037   1.288628    -0.64   0.523    -3.350242    1.704435
              |
  campaignint |   1.788112   .8384941     2.13   0.033     .1436045     3.43262
       polint |   1.048917   1.015529     1.03   0.302    -.9428027    3.040637
      polknow |   .0277809   .5919877     0.05   0.963    -1.133263    1.188825
       gender |  -.1509171   .3149821    -0.48   0.632    -.7686799    .4668456
        White |  -.6334908   .3341754    -1.90   0.058    -1.288897    .0219152
       latinx |  -.3871859   .5541162    -0.70   0.485    -1.473954    .6995818
         ageN |  -.9807477   .9266704    -1.06   0.290    -2.798192     .836697
    education |   2.609106   1.143884     2.28   0.023     .3656486    4.852562
income_norm01 |  -.3938606   .5119075    -0.77   0.442    -1.397846    .6101247
        _cons |  -6.943333   1.467367    -4.73   0.000    -9.821227    -4.06544
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file4D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,791
Number of PSUs   = 1,828                          Population size = 1,599.9581
                                                  Subpop. no. obs =      1,566
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,827

Expression: Pr(volunteer), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .0337486   .0380437     0.89   0.375    -.0408651    .1083623
          2  |   .0077561   .0241543     0.32   0.748    -.0396169    .0551291
------------------------------------------------------------------------------

. svy: logit donatecand c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,826
Number of PSUs   = 1,826                          Population size = 1,598.1421
                                                  Design df       =      1,825
                                                  F(12, 1814)     =       6.94
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .9355193   .4958923     1.89   0.059    -.0370567    1.908095
        1.web |  -.4141821   .5658933    -0.73   0.464    -1.524049    .6956844
              |
  web#c.anger |
           1  |  -.0593617   .6823645    -0.09   0.931    -1.397659    1.278936
              |
  campaignint |   .9883632   .6421246     1.54   0.124     -.271013    2.247739
       polint |   1.275356   .6122253     2.08   0.037     .0746196    2.476092
      polknow |   1.732565   .4209387     4.12   0.000     .9069924    2.558137
       gender |  -.1328539   .1787536    -0.74   0.457    -.4834371    .2177292
        White |  -.0676015   .2536292    -0.27   0.790    -.5650355    .4298324
       latinx |  -.4243581   .3762444    -1.13   0.260    -1.162273    .3135567
         ageN |   .4787991    .681584     0.70   0.482    -.8579675    1.815566
    education |   .7228511    .887806     0.81   0.416    -1.018371    2.464074
income_norm01 |    .439269   .4084298     1.08   0.282      -.36177    1.240308
        _cons |  -5.615812   .8636241    -6.50   0.000    -7.309608   -3.922017
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file5D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,789
Number of PSUs   = 1,826                          Population size = 1,598.1421
                                                  Subpop. no. obs =      1,564
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,825

Expression: Pr(donatecand), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .1219149   .0675107     1.81   0.071    -.0104914    .2543212
          2  |   .0897061   .0488716     1.84   0.067    -.0061442    .1855563
------------------------------------------------------------------------------

. 
. combomarginsplot "file5D" "file4D" "file3D" "file2D" "file1D", labels("Donate" "Volunteer" "Display sign" "Attend rally" "Persuade others") recast(scatter) x(_
> filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("Marginal Effect of Anger by Mode among Dems") title("") legend(pos(6) row(1)) xsize
> (8) ysize(6) xlabel(-.6(.2).6) plot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) ci2op
> ts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(anger16Dem)

  Variables that uniquely identify margins: web _filenumber

. 
. svy: logit persuade c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,613
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Design df       =      1,612
                                                  F(12, 1601)     =       7.75
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.170308   .3841668     3.05   0.002     .4167895    1.923827
        1.web |   1.203244   .3335697     3.61   0.000     .5489678     1.85752
              |
  web#c.anger |
           1  |  -1.102719   .4649879    -2.37   0.018    -2.014764    -.190675
              |
  campaignint |   1.207343    .261709     4.61   0.000     .6940179    1.720669
       polint |   .5916139    .344933     1.72   0.087    -.0849504    1.268178
      polknow |   .2762146   .2547577     1.08   0.278    -.2234764    .7759057
       gender |  -.2123092   .1390127    -1.53   0.127    -.4849738    .0603554
        White |   .2122185    .271647     0.78   0.435    -.3205999    .7450369
       latinx |   .2325076   .3975542     0.58   0.559    -.5472698    1.012285
         ageN |   .3617656   .4321345     0.84   0.403     -.485839     1.20937
    education |  -.3122208   .5595363    -0.56   0.577    -1.409716    .7852743
income_norm01 |  -.0884856    .264367    -0.33   0.738    -.6070248    .4300536
        _cons |  -2.468228   .5225004    -4.72   0.000     -3.49308   -1.443377
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file1R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,576
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Subpop. no. obs =      1,364
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,612

Expression: Pr(persuade), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .2561947   .0796945     3.21   0.001      .099879    .4125104
          2  |   .0154432   .0610592     0.25   0.800    -.1043206    .1352071
------------------------------------------------------------------------------

. svy: logit rally c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,613
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Design df       =      1,612
                                                  F(12, 1601)     =       1.57
                                                  Prob > F        =     0.0942

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .5906305   .7834546     0.75   0.451    -.9460661    2.127327
        1.web |   .5566605   .7881005     0.71   0.480    -.9891487     2.10247
              |
  web#c.anger |
           1  |  -.4013125   .9754002    -0.41   0.681    -2.314498    1.511873
              |
  campaignint |   .6122251   .4996289     1.23   0.221    -.3677654    1.592216
       polint |   .5289772   .7362157     0.72   0.473    -.9150632    1.973018
      polknow |  -.2550917   .5406461    -0.47   0.637    -1.315535    .8053514
       gender |   .0419932   .2774446     0.15   0.880    -.5021969    .5861832
        White |  -.4074962   .5101979    -0.80   0.425    -1.408217    .5932246
       latinx |  -.4299135   .8245133    -0.52   0.602    -2.047144    1.187317
         ageN |  -1.582612    .883335    -1.79   0.073    -3.315218    .1499935
    education |   .6902233    .945015     0.73   0.465    -1.163364     2.54381
income_norm01 |     .86706   .6384407     1.36   0.175     -.385201    2.119321
        _cons |  -4.020338   1.264914    -3.18   0.002    -6.501386   -1.539289
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file2R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,576
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Subpop. no. obs =      1,364
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,612

Expression: Pr(rally), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .0258621   .0355074     0.73   0.467    -.0437835    .0955077
          2  |   .0105635   .0334132     0.32   0.752    -.0549744    .0761014
------------------------------------------------------------------------------

. svy: logit button c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,613
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Design df       =      1,612
                                                  F(12, 1601)     =       4.27
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .7856771    .587369     1.34   0.181    -.3664101    1.937764
        1.web |   .1904753   .5716522     0.33   0.739    -.9307842    1.311735
              |
  web#c.anger |
           1  |   .5369527   .7176693     0.75   0.454    -.8707102    1.944616
              |
  campaignint |   .3022351   .4666523     0.65   0.517    -.6130738    1.217544
       polint |   1.273814   .6038898     2.11   0.035     .0893225    2.458306
      polknow |   .1210478   .3695833     0.33   0.743    -.6038665     .845962
       gender |  -.1691369   .2118556    -0.80   0.425    -.5846783    .2464045
        White |   .4373596   .4293709     1.02   0.309    -.4048241    1.279543
       latinx |   .7609857   .5669352     1.34   0.180    -.3510218    1.872993
         ageN |  -1.560275   .6631451    -2.35   0.019    -2.860992   -.2595574
    education |  -1.293142   .7094509    -1.82   0.069    -2.684685    .0984014
income_norm01 |  -1.181941    .383586    -3.08   0.002    -1.934321   -.4295612
        _cons |  -2.256299   .7573623    -2.98   0.003    -3.741818   -.7707809
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file3R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,576
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Subpop. no. obs =      1,364
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,612

Expression: Pr(button), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .0553164   .0434059     1.27   0.203    -.0298216    .1404544
          2  |   .1419742   .0468109     3.03   0.002     .0501576    .2337908
------------------------------------------------------------------------------

. svy: logit volunteer c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,613
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Design df       =      1,612
                                                  F(12, 1601)     =       0.97
                                                  Prob > F        =     0.4783

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .6667652   .9788358     0.68   0.496    -1.253159     2.58669
        1.web |   .5176289   .9078005     0.57   0.569    -1.262964    2.298222
              |
  web#c.anger |
           1  |  -.3487086    1.20726    -0.29   0.773    -2.716672    2.019255
              |
  campaignint |   .6802386    .769698     0.88   0.377    -.8294753    2.189953
       polint |  -.1463822   1.070624    -0.14   0.891    -2.246344     1.95358
      polknow |   .6052565   .6694516     0.90   0.366    -.7078304    1.918343
       gender |  -.2378504   .3960012    -0.60   0.548    -1.014582     .538881
        White |   .3247211   .7284349     0.45   0.656    -1.104058      1.7535
       latinx |   .1389287   1.165534     0.12   0.905    -2.147192     2.42505
         ageN |  -1.722597   1.330235    -1.29   0.196     -4.33177    .8865746
    education |  -.1899583   1.445321    -0.13   0.895    -3.024863    2.644947
income_norm01 |  -.1362162   .9114509    -0.15   0.881    -1.923969    1.651537
        _cons |  -4.199323   1.806332    -2.32   0.020    -7.742328    -.656318
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file4R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,576
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Subpop. no. obs =      1,364
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,612

Expression: Pr(volunteer), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .0136967   .0223305     0.61   0.540    -.0301033    .0574966
          2  |   .0084228   .0231667     0.36   0.716    -.0370172    .0538627
------------------------------------------------------------------------------

. svy: logit donatecand c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,612
Number of PSUs   = 1,612                          Population size = 1,332.7643
                                                  Design df       =      1,611
                                                  F(12, 1600)     =       5.46
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   2.699537   .8157109     3.31   0.001     1.099571    4.299504
        1.web |    .860183   .8009987     1.07   0.283    -.7109259    2.431292
              |
  web#c.anger |
           1  |  -1.015479   .9512415    -1.07   0.286     -2.88128    .8503217
              |
  campaignint |     1.8651   .6369025     2.93   0.003     .6158554    3.114345
       polint |  -.3216029   .5722263    -0.56   0.574    -1.443989    .8007833
      polknow |   .9106442   .3738442     2.44   0.015     .1773722    1.643916
       gender |   .0066229   .2161572     0.03   0.976    -.4173558    .4306017
        White |  -.7602173   .3815185    -1.99   0.046    -1.508542   -.0118925
       latinx |   .2630985   .5447382     0.48   0.629    -.8053715    1.331568
         ageN |   3.730941   .9001559     4.14   0.000     1.965341    5.496541
    education |    1.66854   .7555538     2.21   0.027     .1865684    3.150512
income_norm01 |  -.0379571   .4868347    -0.08   0.938    -.9928529    .9169388
        _cons |  -8.616266   1.416463    -6.08   0.000    -11.39457   -5.837962
-------------------------------------------------------------------------------

. margins, dydx(anger) at(web=(0 1)) saving("file5R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,575
Number of PSUs   = 1,612                          Population size = 1,332.7643
                                                  Subpop. no. obs =      1,363
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,611

Expression: Pr(donatecand), predict()
dy/dx wrt:  anger
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
anger        |
         _at |
          1  |   .2027421   .0661476     3.06   0.002     .0729978    .3324864
          2  |   .1338354   .0398528     3.36   0.001     .0556667    .2120041
------------------------------------------------------------------------------

. 
. combomarginsplot "file5R" "file4R" "file3R" "file2R" "file1R", labels("Donate" "Volunteer" "Display sign" "Attend rally" "Persuade others") recast(scatter) x(_
> filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("Marginal Effect of Anger by Mode among Reps") title("") legend(pos(6) row(1)) xsize
> (8) ysize(6) xlabel(-.6(.2).6) plot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) ci2op
> ts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(anger16Rep)

  Variables that uniquely identify margins: web _filenumber

. 
. 
. svy: logit persuade c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,826
Number of PSUs   = 1,826                          Population size = 1,600.1414
                                                  Design df       =      1,825
                                                  F(12, 1814)     =       9.62
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .1628709    .343895     0.47   0.636    -.5115983      .83734
        1.web |  -.3710744   .3059882    -1.21   0.225    -.9711982    .2290494
              |
   web#c.fear |
           1  |   .5664959   .4165139     1.36   0.174    -.2503981     1.38339
              |
  campaignint |   1.386682   .2667208     5.20   0.000     .8635724    1.909792
       polint |   .5986223   .3305739     1.81   0.070    -.0497205    1.246965
      polknow |    .748512   .2585077     2.90   0.004     .2415101    1.255514
       gender |   .1306609   .1386928     0.94   0.346    -.1413524    .4026741
        White |   .2768747   .1579618     1.75   0.080    -.0329301    .5866796
       latinx |   .1205489   .2312038     0.52   0.602     -.332903    .5740007
         ageN |  -1.199648   .4111033    -2.92   0.004    -2.005931    -.393366
    education |   .4418515   .4933743     0.90   0.371    -.5257861    1.409489
income_norm01 |   .0255659   .2590986     0.10   0.921    -.4825951    .5337269
        _cons |  -1.738449    .421223    -4.13   0.000    -2.564579   -.9123194
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file6D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,788
Number of PSUs   = 1,826                          Population size = 1,600.1414
                                                  Subpop. no. obs =      1,564
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,825

Expression: Pr(persuade), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .0363875   .0765948     0.48   0.635    -.1138352    .1866102
          2  |   .1598121   .0522121     3.06   0.002     .0574103    .2622139
------------------------------------------------------------------------------

. svy: logit rally c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,825
Number of PSUs   = 1,825                          Population size = 1,599.1822
                                                  Design df       =      1,824
                                                  F(12, 1813)     =       3.73
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |  -.8024842   .4907494    -1.64   0.102    -1.764974    .1600056
        1.web |  -.8203993   .5191344    -1.58   0.114     -1.83856    .1977609
              |
   web#c.fear |
           1  |   1.068008   .6755188     1.58   0.114    -.2568635     2.39288
              |
  campaignint |   1.554703   .5762071     2.70   0.007     .4246079    2.684798
       polint |    .908809   .7098175     1.28   0.201    -.4833316     2.30095
      polknow |   .1123293   .4165454     0.27   0.787    -.7046268    .9292854
       gender |   .0692554   .2169179     0.32   0.750    -.3561782     .494689
        White |   .1720721   .2879798     0.60   0.550    -.3927328    .7368771
       latinx |    .018659   .4265693     0.04   0.965    -.8179567    .8552746
         ageN |   -2.18737   .7161625    -3.05   0.002    -3.591955   -.7827854
    education |   .8218328   .8850687     0.93   0.353    -.9140218    2.557687
income_norm01 |  -.4204728   .4238552    -0.99   0.321    -1.251765    .4108197
        _cons |  -3.085016   .7326751    -4.21   0.000    -4.521987   -1.648046
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file7D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,787
Number of PSUs   = 1,825                          Population size = 1,599.1822
                                                  Subpop. no. obs =      1,563
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,824

Expression: Pr(rally), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |  -.0668935   .0412056    -1.62   0.105    -.1477086    .0139216
          2  |   .0205388   .0346247     0.59   0.553    -.0473694    .0884471
------------------------------------------------------------------------------

. svy: logit button c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,826
Number of PSUs   = 1,826                          Population size = 1,600.1414
                                                  Design df       =      1,825
                                                  F(12, 1814)     =       4.58
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .3393747   .5150632     0.66   0.510    -.6708006     1.34955
        1.web |   .5720679   .4672332     1.22   0.221       -.3443    1.488436
              |
   web#c.fear |
           1  |  -.3508396   .6121926    -0.57   0.567    -1.551511    .8498321
              |
  campaignint |   1.487301   .3861232     3.85   0.000     .7300112    2.244591
       polint |    .908844   .4955284     1.83   0.067    -.0630184    1.880706
      polknow |    .634835   .3159503     2.01   0.045     .0151729    1.254497
       gender |    .088695   .1967292     0.45   0.652     -.297143     .474533
        White |  -.1205644   .2266235    -0.53   0.595     -.565033    .3239041
       latinx |  -.4216757   .3360251    -1.25   0.210     -1.08071    .2373584
         ageN |   -1.51628    .593039    -2.56   0.011    -2.679386    -.353173
    education |  -.5891692   .6376173    -0.92   0.356    -1.839706    .6613671
income_norm01 |   -.071978   .3328138    -0.22   0.829     -.724714    .5807581
        _cons |  -3.135613   .6952657    -4.51   0.000    -4.499213   -1.772012
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file8D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,788
Number of PSUs   = 1,826                          Population size = 1,600.1414
                                                  Subpop. no. obs =      1,564
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,825

Expression: Pr(button), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .0346146   .0531397     0.65   0.515    -.0696063    .1388356
          2  |  -.0014516   .0452191    -0.03   0.974    -.0901383    .0872351
------------------------------------------------------------------------------

. svy: logit volunteer c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,826
Number of PSUs   = 1,826                          Population size = 1,600.1414
                                                  Design df       =      1,825
                                                  F(12, 1814)     =       4.10
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |  -.8918624   .7228314    -1.23   0.217    -2.309526    .5258014
        1.web |  -.0722712   .7171733    -0.10   0.920    -1.478838    1.334296
              |
   web#c.fear |
           1  |   .5242943   .9165081     0.57   0.567    -1.273221    2.321809
              |
  campaignint |   1.965854   .8272733     2.38   0.018     .3433517    3.588356
       polint |   1.093535   1.010435     1.08   0.279    -.8881963    3.075265
      polknow |    .025179   .5915375     0.04   0.966    -1.134983    1.185341
       gender |  -.0691593   .3186991    -0.22   0.828    -.6942126     .555894
        White |  -.6622941   .3366228    -1.97   0.049      -1.3225   -.0020877
       latinx |  -.3549729   .5529294    -0.64   0.521    -1.439414    .7294681
         ageN |  -1.064183   .9033844    -1.18   0.239    -2.835959    .7075928
    education |   2.823805    1.15192     2.45   0.014     .5645858    5.083024
income_norm01 |  -.3660128   .5169193    -0.71   0.479    -1.379828    .6478027
        _cons |  -5.982356   1.279305    -4.68   0.000    -8.491411     -3.4733
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file9D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,788
Number of PSUs   = 1,826                          Population size = 1,600.1414
                                                  Subpop. no. obs =      1,564
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,825

Expression: Pr(volunteer), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   -.026887   .0231568    -1.16   0.246    -.0723036    .0185297
          2  |  -.0141115   .0195763    -0.72   0.471    -.0525057    .0242828
------------------------------------------------------------------------------

. svy: logit donatecand c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,824
Number of PSUs   = 1,824                          Population size = 1,598.3254
                                                  Design df       =      1,823
                                                  F(12, 1812)     =       6.59
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .4758779    .507859     0.94   0.349    -.5201687    1.471924
        1.web |  -.5297482   .5670001    -0.93   0.350    -1.641786    .5822899
              |
   web#c.fear |
           1  |   .1370395   .6725764     0.20   0.839    -1.182062    1.456141
              |
  campaignint |   1.048958   .6529832     1.61   0.108    -.2317159    2.329632
       polint |   1.330565   .6200133     2.15   0.032     .1145542    2.546576
      polknow |   1.752435   .4318842     4.06   0.000     .9053951    2.599475
       gender |  -.1111781   .1801566    -0.62   0.537    -.4645131    .2421568
        White |  -.0664107   .2533894    -0.26   0.793    -.5633748    .4305533
       latinx |  -.3760333   .3755124    -1.00   0.317    -1.112513    .3604464
         ageN |   .3154714   .6829745     0.46   0.644    -1.024023    1.654966
    education |   .7452502   .8786634     0.85   0.396    -.9780425    2.468543
income_norm01 |   .4431423   .4042088     1.10   0.273    -.3496188    1.235903
        _cons |  -5.337958   .9093833    -5.87   0.000      -7.1215   -3.554415
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file10D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,786
Number of PSUs   = 1,824                          Population size = 1,598.3254
                                                  Subpop. no. obs =      1,562
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,823

Expression: Pr(donatecand), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .0617589   .0657772     0.94   0.348    -.0672476    .1907654
          2  |   .0631167   .0412686     1.53   0.126     -.017822    .1440554
------------------------------------------------------------------------------

. 
. combomarginsplot "file10D" "file9D" "file8D" "file7D" "file6D", labels("Donate" "Volunteer" "Display sign" "Attend rally" "Persuade others") recast(scatter) x(
> _filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("Marginal Effect of Fear by Mode among Dems") title("") legend(pos(6) row(1)) xsize
> (8) ysize(6) xlabel(-.6(.2).6) plot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) ci2op
> ts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(fear16Dem)

  Variables that uniquely identify margins: web _filenumber

. 
. svy: logit persuade c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      8.34
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.440056   .3545849     4.06   0.000     .7445612    2.135551
        1.web |   1.323541   .3004719     4.40   0.000     .7341853    1.912897
              |
   web#c.fear |
           1  |  -1.384068   .4252257    -3.25   0.001     -2.21812   -.5500161
              |
  campaignint |   1.237032   .2655432     4.66   0.000     .7161868    1.757878
       polint |   .5626144   .3495745     1.61   0.108    -.1230529    1.248282
      polknow |   .2976264   .2552475     1.17   0.244    -.2030246    .7982775
       gender |  -.2101915   .1398728    -1.50   0.133    -.4845428    .0641597
        White |   .2144418    .274638     0.78   0.435    -.3242425    .7531262
       latinx |   .2561236   .4016333     0.64   0.524    -.5316537    1.043901
         ageN |   .3066391   .4351804     0.70   0.481    -.5469386    1.160217
    education |  -.3176822   .5564727    -0.57   0.568    -1.409167    .7738023
income_norm01 |  -.0760086   .2648773    -0.29   0.774    -.5955479    .4435306
        _cons |  -2.574688    .522586    -4.93   0.000    -3.599706    -1.54967
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file6R")

Average marginal effects

Number of strata =     1                           Number of obs   =     1,579
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Subpop. no. obs =     1,367
                                                   Subpop. size    =         .
Model VCE: Linearized                              Design df       =     1,615

Expression: Pr(persuade), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .3066296   .0682136     4.50   0.000     .1728331    .4404261
          2  |   .0127898   .0546605     0.23   0.815    -.0944231    .1200028
------------------------------------------------------------------------------

. svy: logit rally c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      1.50
                                                   Prob > F        =    0.1165

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .0921361   .9200882     0.10   0.920    -1.712556    1.896828
        1.web |   .2927792   .7748356     0.38   0.706     -1.22701    1.812568
              |
   web#c.fear |
           1  |   .0148811   1.087616     0.01   0.989    -2.118406    2.148168
              |
  campaignint |   .6361482   .4996852     1.27   0.203    -.3439513    1.616248
       polint |   .5490428   .7301141     0.75   0.452    -.8830278    1.981113
      polknow |  -.2567544    .549119    -0.47   0.640    -1.333815    .8203062
       gender |   .0489501    .279662     0.18   0.861    -.4995885    .5974886
        White |  -.3939407   .5050408    -0.78   0.435    -1.384545    .5966635
       latinx |  -.4379307   .8286753    -0.53   0.597    -2.063323    1.187461
         ageN |  -1.599103    .879427    -1.82   0.069    -3.324041    .1258346
    education |   .6304133   .9540976     0.66   0.509    -1.240986    2.501813
income_norm01 |   .8841622   .6296414     1.40   0.160    -.3508378    2.119162
        _cons |  -3.702044   1.415914    -2.61   0.009    -6.479266   -.9248226
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file7R")

Average marginal effects

Number of strata =     1                           Number of obs   =     1,579
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Subpop. no. obs =     1,367
                                                   Subpop. size    =         .
Model VCE: Linearized                              Design df       =     1,615

Expression: Pr(rally), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .0039505   .0395324     0.10   0.920    -.0735896    .0814906
          2  |    .005977   .0323077     0.19   0.853    -.0573925    .0693465
------------------------------------------------------------------------------

. svy: logit button c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      3.61
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .8738197   .5558649     1.57   0.116    -.2164725    1.964112
        1.web |   .8250435   .5020219     1.64   0.100    -.1596392    1.809726
              |
   web#c.fear |
           1  |  -.3578268   .6663928    -0.54   0.591    -1.664912    .9492586
              |
  campaignint |   .3061883    .459532     0.67   0.505    -.5951534     1.20753
       polint |   1.319697   .6027726     2.19   0.029     .1373981    2.501996
      polknow |   .1041775   .3746267     0.28   0.781    -.6306279     .838983
       gender |  -.1633819   .2114499    -0.77   0.440    -.5781269    .2513632
        White |   .5781665   .4358842     1.33   0.185    -.2767916    1.433125
       latinx |   .7488249   .5581401     1.34   0.180      -.34593     1.84358
         ageN |  -1.670504   .6693307    -2.50   0.013    -2.983352   -.3576559
    education |  -1.292187     .71894    -1.80   0.072     -2.70234    .1179664
income_norm01 |  -1.111953    .380286    -2.92   0.004    -1.857859   -.3660468
        _cons |   -2.39417   .7607274    -3.15   0.002    -3.886287   -.9020539
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file8R")

Average marginal effects

Number of strata =     1                           Number of obs   =     1,579
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Subpop. no. obs =     1,367
                                                   Subpop. size    =         .
Model VCE: Linearized                              Design df       =     1,615

Expression: Pr(button), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .0626529   .0429224     1.46   0.145    -.0215366    .1468424
          2  |   .0568122   .0433145     1.31   0.190    -.0281463    .1417706
------------------------------------------------------------------------------

. svy: logit volunteer c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      1.29
                                                   Prob > F        =    0.2181

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.144407   .7823453     1.46   0.144    -.3901112    2.678926
        1.web |   .6044782   .8776678     0.69   0.491    -1.117009    2.325966
              |
   web#c.fear |
           1  |  -.5107372   1.026816    -0.50   0.619     -2.52477    1.503295
              |
  campaignint |   .6723823   .7691365     0.87   0.382    -.8362283    2.180993
       polint |  -.2400205   1.099079    -0.22   0.827    -2.395791    1.915751
      polknow |   .6318279   .6762711     0.93   0.350    -.6946332    1.958289
       gender |  -.2432261    .396362    -0.61   0.540    -1.020664    .5342117
        White |   .2927965    .733259     0.40   0.690    -1.145443    1.731036
       latinx |   .1450015   1.176849     0.12   0.902    -2.163309    2.453312
         ageN |  -1.754872   1.322602    -1.33   0.185    -4.349068    .8393245
    education |  -.1309392   1.432395    -0.09   0.927    -2.940488    2.678609
income_norm01 |  -.0914592   .8905761    -0.10   0.918    -1.838265    1.655347
        _cons |  -4.454852   1.712426    -2.60   0.009    -7.813662   -1.096042
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file9R")

Average marginal effects

Number of strata =     1                           Number of obs   =     1,579
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Subpop. no. obs =     1,367
                                                   Subpop. size    =         .
Model VCE: Linearized                              Design df       =     1,615

Expression: Pr(volunteer), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .0238816   .0192084     1.24   0.214    -.0137944    .0615575
          2  |   .0166578   .0224922     0.74   0.459    -.0274591    .0607747
------------------------------------------------------------------------------

. svy: logit donatecand c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,334.0214
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       4.63
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.573378   .7296592     2.16   0.031      .142199    3.004557
        1.web |   .3062433   .6841214     0.45   0.654    -1.035616    1.648103
              |
   web#c.fear |
           1  |  -.3214345   .8358451    -0.38   0.701     -1.96089    1.318021
              |
  campaignint |   1.841326   .6240515     2.95   0.003     .6172898    3.065363
       polint |  -.2608015    .574542    -0.45   0.650    -1.387728    .8661253
      polknow |   .9505059   .3780234     2.51   0.012     .2090376    1.691974
       gender |   .0061179    .216447     0.03   0.977    -.4184288    .4306646
        White |  -.6942795   .3687406    -1.88   0.060     -1.41754    .0289812
       latinx |   .2082203   .5432698     0.38   0.702    -.8573681    1.273809
         ageN |   3.654431   .9004476     4.06   0.000     1.888262    5.420601
    education |   1.683351   .7590775     2.22   0.027     .1944695    3.172232
income_norm01 |   .0583352   .4785026     0.12   0.903    -.8802166    .9968869
        _cons |  -7.810195   1.381165    -5.65   0.000    -10.51926    -5.10113
-------------------------------------------------------------------------------

. margins, dydx(fear) at(web=(0 1)) saving("file10R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,615                          Population size = 1,334.0214
                                                  Subpop. no. obs =      1,366
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,614

Expression: Pr(donatecand), predict()
dy/dx wrt:  fear
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
fear         |
         _at |
          1  |   .1201851   .0557564     2.16   0.031     .0108225    .2295477
          2  |   .1008452   .0366124     2.75   0.006     .0290324    .1726579
------------------------------------------------------------------------------

. 
. combomarginsplot "file10R" "file9R" "file8R" "file7R" "file6R", labels("Donate" "Volunteer" "Display sign" "Attend rally" "Persuade others") recast(scatter) x(
> _filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("Marginal Effect of Fear by Mode among Reps") title("") legend(pos(6) row(1)) xsize
> (8) ysize(6) xlabel(-.6(.2).6) plot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) ci2op
> ts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(fear16Rep)

  Variables that uniquely identify margins: web _filenumber

. 
. 
. svy: logit persuade c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,601.3724
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       9.33
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   .1361064   .4260928     0.32   0.749    -.6995737    .9717865
        1.web |  -.3032283    .287347    -1.06   0.291    -.8667914    .2603349
              |
web#c.hopeful |
           1  |   .6735055   .4901369     1.37   0.170     -.287782    1.634793
              |
  campaignint |   1.306806   .2634932     4.96   0.000     .7900261    1.823585
       polint |   .5814496   .3324745     1.75   0.080    -.0706204     1.23352
      polknow |   .7220862   .2595895     2.78   0.005     .2129628     1.23121
       gender |    .137729   .1398396     0.98   0.325    -.1365334    .4119914
        White |   .2892653   .1601029     1.81   0.071    -.0247386    .6032692
       latinx |   .0954457   .2359098     0.40   0.686    -.3672356    .5581269
         ageN |  -1.239783   .4196324    -2.95   0.003    -2.062793   -.4167734
    education |   .5244414   .4989881     1.05   0.293    -.4542057    1.503088
income_norm01 |   .0814906   .2622189     0.31   0.756    -.4327896    .5957708
        _cons |  -1.705009   .4335536    -3.93   0.000    -2.555322   -.8546964
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file11D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,790
Number of PSUs   = 1,828                          Population size = 1,601.3724
                                                  Subpop. no. obs =      1,566
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,827

Expression: Pr(persuade), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0305811   .0954669     0.32   0.749    -.1566545    .2178168
          2  |   .1771842   .0553912     3.20   0.001     .0685474    .2858209
------------------------------------------------------------------------------

. svy: logit rally c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,827
Number of PSUs   = 1,827                          Population size = 1,600.4132
                                                  Design df       =      1,826
                                                  F(12, 1815)     =       3.64
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |    .269205   .6016896     0.45   0.655    -.9108671    1.449277
        1.web |   -.501993   .4709812    -1.07   0.287    -1.425711    .4217254
              |
web#c.hopeful |
           1  |   .6099212   .7227011     0.84   0.399    -.8074865    2.027329
              |
  campaignint |   1.348704   .5668378     2.38   0.017     .2369856    2.460423
       polint |   .8700613   .7054642     1.23   0.218    -.5135401    2.253663
      polknow |    .029332    .412461     0.07   0.943     -.779613    .8382769
       gender |   .0102101   .2130974     0.05   0.962    -.4077301    .4281503
        White |   .2526991   .2864712     0.88   0.378    -.3091465    .8145448
       latinx |  -.0104218   .4258905    -0.02   0.980    -.8457056    .8248619
         ageN |  -2.263524   .7266112    -3.12   0.002      -3.6886   -.8384472
    education |   .7743009   .8803316     0.88   0.379    -.9522617    2.500864
income_norm01 |  -.3958382   .4215935    -0.94   0.348    -1.222694    .4310178
        _cons |  -3.436698   .7188946    -4.78   0.000     -4.84664   -2.026756
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file12D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,789
Number of PSUs   = 1,827                          Population size = 1,600.4132
                                                  Subpop. no. obs =      1,565
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,826

Expression: Pr(rally), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0233418    .052277     0.45   0.655    -.0791871    .1258707
          2  |    .067821   .0338222     2.01   0.045     .0014868    .1341553
------------------------------------------------------------------------------

. svy: logit button c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,601.3724
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       5.72
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   .8293821   .6662628     1.24   0.213    -.4773346    2.136099
        1.web |   .1258311    .478871     0.26   0.793     -.813361    1.065023
              |
web#c.hopeful |
           1  |   .3466657   .7388875     0.47   0.639    -1.102487    1.795819
              |
  campaignint |   1.268425   .3676872     3.45   0.001     .5472939    1.989557
       polint |    .844476    .476834     1.77   0.077    -.0907211    1.779673
      polknow |   .5818281   .3198578     1.82   0.069    -.0454974    1.209154
       gender |   .0537348    .195809     0.27   0.784    -.3302982    .4377677
        White |  -.0105029   .2249023    -0.05   0.963    -.4515955    .4305897
       latinx |  -.4230612   .3334575    -1.27   0.205    -1.077059    .2309369
         ageN |  -1.709091   .6015349    -2.84   0.005    -2.888859   -.5293221
    education |  -.5191718   .6328139    -0.82   0.412    -1.760286    .7219429
income_norm01 |  -.0921781   .3429373    -0.27   0.788    -.7647684    .5804122
        _cons |  -3.127941   .6727051    -4.65   0.000    -4.447292   -1.808589
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file13D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,790
Number of PSUs   = 1,828                          Population size = 1,601.3724
                                                  Subpop. no. obs =      1,566
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,827

Expression: Pr(button), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0833626   .0676121     1.23   0.218    -.0492426    .2159678
          2  |   .1456163   .0434561     3.35   0.001     .0603875    .2308452
------------------------------------------------------------------------------

. svy: logit volunteer c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,601.3724
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       5.03
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |    1.51993   1.099976     1.38   0.167     -.637413    3.677274
        1.web |   .4528109   .8951097     0.51   0.613    -1.302735    2.208357
              |
web#c.hopeful |
           1  |  -.3946697   1.269439    -0.31   0.756    -2.884374    2.095035
              |
  campaignint |   1.595283   .8279831     1.93   0.054    -.0286095    3.219176
       polint |   1.002481   1.013065     0.99   0.323    -.9844064    2.989368
      polknow |  -.0479273    .598585    -0.08   0.936     -1.22191    1.126056
       gender |  -.1966992   .3138286    -0.63   0.531    -.8121998    .4188014
        White |  -.5404239   .3219328    -1.68   0.093    -1.171819    .0909711
       latinx |  -.3678141   .5469851    -0.67   0.501    -1.440596    .7049678
         ageN |   -1.27079   .9093817    -1.40   0.162    -3.054327    .5127467
    education |   2.838672   1.130153     2.51   0.012     .6221452    5.055199
income_norm01 |  -.4149569   .5301813    -0.78   0.434    -1.454782    .6248682
        _cons |  -6.919331   1.363436    -5.07   0.000    -9.593388   -4.245275
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file14D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,790
Number of PSUs   = 1,828                          Population size = 1,601.3724
                                                  Subpop. no. obs =      1,566
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,827

Expression: Pr(volunteer), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |    .048491   .0382997     1.27   0.206    -.0266248    .1236068
          2  |   .0424563   .0240211     1.77   0.077    -.0046553     .089568
------------------------------------------------------------------------------

. svy: logit donatecand c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,826
Number of PSUs   = 1,826                          Population size = 1,599.5564
                                                  Design df       =      1,825
                                                  F(12, 1814)     =       6.65
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   .7816183   .6116851     1.28   0.201    -.4180581    1.981295
        1.web |    -.52328   .5473059    -0.96   0.339    -1.596692    .5501317
              |
web#c.hopeful |
           1  |   .2265315   .7880191     0.29   0.774    -1.318983    1.772046
              |
  campaignint |    .890789   .6317324     1.41   0.159    -.3482056    2.129784
       polint |   1.274628    .613812     2.08   0.038       .07078    2.478476
      polknow |   1.693979   .4263728     3.97   0.000     .8577487    2.530209
       gender |  -.1221725   .1761165    -0.69   0.488    -.4675836    .2232386
        White |  -.0314484   .2485604    -0.13   0.899    -.5189412    .4560444
       latinx |  -.3998073   .3792274    -1.05   0.292    -1.143573     .343958
         ageN |   .1904025   .6927301     0.27   0.783    -1.168225     1.54903
    education |    .931708   .8676572     1.07   0.283    -.7699974    2.633413
income_norm01 |   .4603792   .4073433     1.13   0.259    -.3385289    1.259287
        _cons |  -5.362785   .9256091    -5.79   0.000    -7.178149    -3.54742
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file15D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,788
Number of PSUs   = 1,826                          Population size = 1,599.5564
                                                  Subpop. no. obs =      1,564
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,825

Expression: Pr(donatecand), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0994068   .0761921     1.30   0.192    -.0500262    .2488397
          2  |   .1036733   .0461808     2.24   0.025     .0131006     .194246
------------------------------------------------------------------------------

. 
. combomarginsplot "file15D" "file14D" "file13D" "file12D" "file11D", labels("Donate" "Volunteer" "Display sign" "Attend rally" "Persuade others") recast(scatter
> ) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("Marginal Effect of Hope by Mode among Dems") title("") legend(pos(6) row(1)) x
> size(8) ysize(6) xlabel(-.6(.2).6) plot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) c
> i2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(hope16Dem)

  Variables that uniquely identify margins: web _filenumber

. 
. svy: logit persuade c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       8.64
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.459982   .4161464     3.51   0.000     .6437381    2.276226
        1.web |   .7188059   .2838777     2.53   0.011     .1619983    1.275614
              |
web#c.hopeful |
           1  |   -.521218   .4854416    -1.07   0.283     -1.47338    .4309441
              |
  campaignint |   1.074166   .2701491     3.98   0.000     .5442866    1.604046
       polint |   .4676157   .3500239     1.34   0.182    -.2189334    1.154165
      polknow |   .3780094   .2608411     1.45   0.147    -.1336134    .8896322
       gender |  -.1724494   .1411714    -1.22   0.222    -.4493479    .1044491
        White |    .131241   .2916823     0.45   0.653     -.440875    .7033569
       latinx |   .2532168   .4223173     0.60   0.549     -.575131    1.081565
         ageN |   .2319178   .4430294     0.52   0.601    -.6370555    1.100891
    education |  -.0388122   .5704184    -0.07   0.946    -1.157651    1.080026
income_norm01 |   .0107941   .2666961     0.04   0.968     -.512313    .5339012
        _cons |  -2.430441   .5231853    -4.65   0.000    -3.456634   -1.404247
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file11R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Subpop. no. obs =      1,366
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,614

Expression: Pr(persuade), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .3156455   .0821964     3.84   0.000     .1544226    .4768684
          2  |    .210744    .058287     3.62   0.000     .0964178    .3250702
------------------------------------------------------------------------------

. svy: logit rally c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       1.72
                                                  Prob > F        =     0.0572

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |    1.68418   .7792795     2.16   0.031     .1556736    3.212686
        1.web |   .6709636   .5999179     1.12   0.264    -.5057362    1.847663
              |
web#c.hopeful |
           1  |  -.7558446   .9049971    -0.84   0.404    -2.530938    1.019248
              |
  campaignint |    .454625   .5143869     0.88   0.377    -.5543114    1.463561
       polint |   .3925192   .7107622     0.55   0.581    -1.001595    1.786633
      polknow |   -.141039   .5495019    -0.26   0.797    -1.218851    .9367732
       gender |   .0857764   .2703107     0.32   0.751    -.4444204    .6159732
        White |  -.4844173   .5063124    -0.96   0.339    -1.477516    .5086816
       latinx |  -.4723906   .8373705    -0.56   0.573    -2.114838    1.170057
         ageN |  -1.750646   .9066714    -1.93   0.054    -3.529023    .0277305
    education |   1.083777   .9855829     1.10   0.272    -.8493793    3.016934
income_norm01 |   .9480187   .6519383     1.45   0.146    -.3307159    2.226753
        _cons |  -4.561626   1.145343    -3.98   0.000    -6.808142   -2.315111
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file12R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Subpop. no. obs =      1,366
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,614

Expression: Pr(rally), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0760381   .0462942     1.64   0.101     -.014765    .1668412
          2  |   .0508887    .030219     1.68   0.092    -.0083839    .1101613
------------------------------------------------------------------------------

. svy: logit button c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       4.20
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.504206   .7263726     2.07   0.039      .079473    2.928938
        1.web |    .499436   .5753836     0.87   0.386    -.6291415    1.628013
              |
web#c.hopeful |
           1  |   .1113768   .8808998     0.13   0.899    -1.616451    1.839204
              |
  campaignint |   .0881319   .4367513     0.20   0.840    -.7685274    .9447912
       polint |   1.152851   .5655995     2.04   0.042      .043464    2.262237
      polknow |   .2256328   .3660683     0.62   0.538    -.4923864     .943652
       gender |  -.1227527   .2108229    -0.58   0.560    -.5362681    .2907628
        White |    .458987   .4300857     1.07   0.286    -.3845981    1.302572
       latinx |   .7691046   .5668711     1.36   0.175    -.3427761    1.880985
         ageN |  -1.789211   .6697624    -2.67   0.008    -3.102906   -.4755152
    education |  -.8897199   .7199739    -1.24   0.217    -2.301902    .5224621
income_norm01 |  -1.064381   .3793637    -2.81   0.005    -1.808478   -.3202838
        _cons |  -2.600078   .7976534    -3.26   0.001    -4.164624   -1.035533
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file13R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Subpop. no. obs =      1,366
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,614

Expression: Pr(button), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .1071305   .0583697     1.84   0.067    -.0073579    .2216188
          2  |   .1719045   .0499292     3.44   0.001     .0739718    .2698373
------------------------------------------------------------------------------

. svy: logit volunteer c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       1.44
                                                  Prob > F        =     0.1408

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.095222   .7778162     1.41   0.159    -.4304138    2.620858
        1.web |   .8468814   .8462851     1.00   0.317    -.8130516    2.506814
              |
web#c.hopeful |
           1  |  -1.043466   1.231863    -0.85   0.397    -3.459684    1.372753
              |
  campaignint |   .6863256   .7081811     0.97   0.333    -.7027256    2.075377
       polint |  -.1726199   1.024477    -0.17   0.866    -2.182064    1.836825
      polknow |   .6355976   .6456333     0.98   0.325      -.63077    1.901965
       gender |  -.2270977   .3911746    -0.58   0.562    -.9943613    .5401658
        White |   .3439404   .7512905     0.46   0.647    -1.129667    1.817548
       latinx |   .1093747   1.165651     0.09   0.925    -2.176973    2.395723
         ageN |  -1.794645    1.33232    -1.35   0.178    -4.407904    .8186135
    education |  -.1927622   1.509517    -0.13   0.898    -3.153581    2.768056
income_norm01 |  -.1159552    .961492    -0.12   0.904    -2.001859    1.769949
        _cons |  -4.332179   1.836635    -2.36   0.018     -7.93462   -.7297392
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file14R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Subpop. no. obs =      1,366
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,614

Expression: Pr(volunteer), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .0226911   .0210334     1.08   0.281    -.0185647    .0639468
          2  |   .0013784   .0238346     0.06   0.954    -.0453715    .0481284
------------------------------------------------------------------------------

. svy: logit donatecand c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,614
Number of PSUs   = 1,614                          Population size = 1,333.5352
                                                  Design df       =      1,613
                                                  F(12, 1602)     =       4.90
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   3.308953   .7320798     4.52   0.000     1.873025     4.74488
        1.web |   1.088338   .6658827     1.63   0.102    -.2177486    2.394424
              |
web#c.hopeful |
           1  |  -1.557768   .9120505    -1.71   0.088    -3.346697    .2311604
              |
  campaignint |   1.525767   .6215417     2.45   0.014     .3066525    2.744881
       polint |  -.3112706   .5418025    -0.57   0.566    -1.373981    .7514401
      polknow |   1.107635   .3638865     3.04   0.002     .3938952    1.821375
       gender |   .0162663   .2189162     0.07   0.941    -.4131237    .4456564
        White |  -.7746093    .381251    -2.03   0.042    -1.522409     -.02681
       latinx |   .1918209   .5178079     0.37   0.711     -.823826    1.207468
         ageN |   3.693247   .9392484     3.93   0.000     1.850972    5.535523
    education |   2.154395   .7894108     2.73   0.006     .6060167    3.702774
income_norm01 |   .0597735   .4988752     0.12   0.905    -.9187381    1.038285
        _cons |  -8.828069   1.423346    -6.20   0.000    -11.61987   -6.036268
-------------------------------------------------------------------------------

. margins, dydx(hopeful) at(web=(0 1)) saving("file15R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,577
Number of PSUs   = 1,614                          Population size = 1,333.5352
                                                  Subpop. no. obs =      1,365
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,613

Expression: Pr(donatecand), predict()
dy/dx wrt:  hopeful
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
hopeful      |
         _at |
          1  |   .2416729   .0590483     4.09   0.000     .1258535    .3574922
          2  |   .1390271   .0395819     3.51   0.000     .0613898    .2166644
------------------------------------------------------------------------------

. 
. combomarginsplot "file15R" "file14R" "file13R" "file12R" "file11R", labels("Donate" "Volunteer" "Display sign" "Attend rally" "Persuade others") recast(scatter
> ) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("Marginal Effect of Hope by Mode among Reps") title("") legend(pos(6) row(1)) x
> size(8) ysize(6) xlabel(-.6(.2).6) plot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) c
> i2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(hope16Rep)

  Variables that uniquely identify margins: web _filenumber

. 
. 
. svy: logit persuade c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,829
Number of PSUs   = 1,829                          Population size = 1,602.0694
                                                  Design df       =      1,828
                                                  F(12, 1817)     =       9.81
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .3607271   .4068033     0.89   0.375    -.4371211    1.158575
        1.web |  -.2496421   .2546311    -0.98   0.327    -.7490404    .2497563
              |
  web#c.proud |
           1  |   .6038168   .4623257     1.31   0.192    -.3029254    1.510559
              |
  campaignint |   1.254036   .2656715     4.72   0.000     .7329847    1.775088
       polint |   .5863403   .3312848     1.77   0.077    -.0633961    1.236077
      polknow |   .6952423   .2539962     2.74   0.006      .197089    1.193396
       gender |     .12055   .1407967     0.86   0.392    -.1555893    .3966893
        White |   .3361357   .1593803     2.11   0.035      .023549    .6487223
       latinx |   .1101428   .2352763     0.47   0.640    -.3512957    .5715813
         ageN |  -1.272959   .4176813    -3.05   0.002    -2.092142   -.4537761
    education |   .5417784   .4969935     1.09   0.276    -.4329564    1.516513
income_norm01 |   .1116151   .2609443     0.43   0.669    -.4001652    .6233954
        _cons |  -1.799246   .4124475    -4.36   0.000    -2.608164   -.9903282
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file16D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,792
Number of PSUs   = 1,829                          Population size = 1,602.0694
                                                  Subpop. no. obs =      1,567
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,828

Expression: Pr(persuade), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .0807302   .0900841     0.90   0.370    -.0959483    .2574087
          2  |   .2097848   .0525061     4.00   0.000     .1068065    .3127631
------------------------------------------------------------------------------

. svy: logit rally c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,601.1102
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       3.38
                                                  Prob > F        =     0.0001

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |  -.0997543   .5760057    -0.17   0.863    -1.229453    1.029944
        1.web |  -.6589086   .4468541    -1.47   0.141    -1.535307    .2174899
              |
  web#c.proud |
           1  |   .9382044   .7010134     1.34   0.181    -.4366675    2.313076
              |
  campaignint |   1.380935   .5664664     2.44   0.015     .2699451    2.491925
       polint |   .8781565    .700083     1.25   0.210    -.4948906    2.251204
      polknow |   .0483596   .4183726     0.12   0.908    -.7721791    .8688984
       gender |   .0103827   .2127526     0.05   0.961    -.4068812    .4276467
        White |    .254961   .2848887     0.89   0.371    -.3037807    .8137026
       latinx |  -.0002529    .425742    -0.00   1.000     -.835245    .8347391
         ageN |  -2.217516   .7315318    -3.03   0.002    -3.652243   -.7827898
    education |   .7327604   .8918659     0.82   0.411    -1.016423    2.481944
income_norm01 |  -.3754031   .4209629    -0.89   0.373    -1.201022    .4502159
        _cons |  -3.285185    .726677    -4.52   0.000     -4.71039   -1.859981
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file17D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,791
Number of PSUs   = 1,828                          Population size = 1,601.1102
                                                  Subpop. no. obs =      1,566
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,827

Expression: Pr(rally), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |  -.0086541   .0500053    -0.17   0.863    -.1067276    .0894194
          2  |    .064835   .0336047     1.93   0.054    -.0010726    .1307426
------------------------------------------------------------------------------

. svy: logit button c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,829
Number of PSUs   = 1,829                          Population size = 1,602.0694
                                                  Design df       =      1,828
                                                  F(12, 1817)     =       5.83
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .4768511    .587723     0.81   0.417     -.675828     1.62953
        1.web |  -.0976409   .4287863    -0.23   0.820    -.9386033    .7433215
              |
  web#c.proud |
           1  |   .7719499   .6669342     1.16   0.247    -.5360832    2.079983
              |
  campaignint |   1.282256   .3757057     3.41   0.001     .5453985    2.019113
       polint |   .8418282   .4762629     1.77   0.077    -.0922484    1.775905
      polknow |   .5599022   .3255165     1.72   0.086    -.0785211    1.198325
       gender |   .0359035   .1957203     0.18   0.854    -.3479554    .4197625
        White |   .0262372   .2281241     0.12   0.908    -.4211741    .4736485
       latinx |  -.4005062   .3351121    -1.20   0.232    -1.057749    .2567366
         ageN |  -1.673786   .6042201    -2.77   0.006     -2.85882   -.4887519
    education |  -.5545857   .6382383    -0.87   0.385    -1.806339    .6971671
income_norm01 |  -.0562145   .3431002    -0.16   0.870    -.7291241    .6166952
        _cons |  -2.945743    .650689    -4.53   0.000    -4.221915   -1.669571
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file18D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,792
Number of PSUs   = 1,829                          Population size = 1,602.0694
                                                  Subpop. no. obs =      1,567
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,828

Expression: Pr(button), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |    .047868   .0591956     0.81   0.419    -.0682302    .1639662
          2  |   .1547287   .0443975     3.49   0.001     .0676536    .2418038
------------------------------------------------------------------------------

. svy: logit volunteer c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,829
Number of PSUs   = 1,829                          Population size = 1,602.0694
                                                  Design df       =      1,828
                                                  F(12, 1817)     =       4.78
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |    .678995   .8245928     0.82   0.410    -.9382479    2.296238
        1.web |   .1033241   .7048444     0.15   0.883    -1.279061    1.485709
              |
  web#c.proud |
           1  |   .1911239   1.022366     0.19   0.852    -1.814003    2.196251
              |
  campaignint |   1.667683   .8349129     2.00   0.046     .0301999    3.305167
       polint |   1.019176    .994903     1.02   0.306    -.9320898    2.970442
      polknow |   -.023572   .6020045    -0.04   0.969    -1.204261    1.157117
       gender |  -.1867359   .3077031    -0.61   0.544    -.7902224    .4167507
        White |  -.5388168   .3301431    -1.63   0.103    -1.186314    .1086805
       latinx |  -.3343066   .5519556    -0.61   0.545    -1.416837    .7482233
         ageN |  -1.179909   .8952638    -1.32   0.188    -2.935756    .5759388
    education |   2.728966   1.148486     2.38   0.018     .4764831    4.981449
income_norm01 |  -.3765916   .5268293    -0.71   0.475    -1.409842     .656659
        _cons |  -6.472019   1.263819    -5.12   0.000    -8.950699   -3.993338
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file19D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,792
Number of PSUs   = 1,829                          Population size = 1,602.0694
                                                  Subpop. no. obs =      1,567
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,828

Expression: Pr(volunteer), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .0213089   .0267541     0.80   0.426    -.0311628    .0737807
          2  |   .0330868   .0224562     1.47   0.141    -.0109557    .0771293
------------------------------------------------------------------------------

. svy: logit donatecand c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,827
Number of PSUs   = 1,827                          Population size = 1,600.2534
                                                  Design df       =      1,826
                                                  F(12, 1815)     =       7.06
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.369704   .5830731     2.35   0.019     .2261435    2.513264
        1.web |  -.2335169   .4778319    -0.49   0.625    -1.170671    .7036376
              |
  web#c.proud |
           1  |  -.1887099   .7149775    -0.26   0.792     -1.59097     1.21355
              |
  campaignint |   .8593972   .6345819     1.35   0.176    -.3851855     2.10398
       polint |    1.21567   .6004422     2.02   0.043      .038044    2.393295
      polknow |   1.613927   .4108253     3.93   0.000     .8081906    2.419664
       gender |   -.174028   .1788531    -0.97   0.331    -.5248061    .1767502
        White |   .0567449   .2502522     0.23   0.821    -.4340658    .5475556
       latinx |  -.3375818   .3783099    -0.89   0.372    -1.079547    .4043838
         ageN |   .1800994   .6886952     0.26   0.794    -1.170614    1.530813
    education |   .9746166    .831023     1.17   0.241    -.6552388    2.604472
income_norm01 |   .4892371   .3997362     1.22   0.221     -.294751    1.273225
        _cons |  -5.686476   .8528982    -6.67   0.000    -7.359234   -4.013717
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file20D")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,790
Number of PSUs   = 1,827                          Population size = 1,600.2534
                                                  Subpop. no. obs =      1,565
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,826

Expression: Pr(donatecand), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .1685674   .0723409     2.33   0.020     .0266878    .3104471
          2  |   .1218408   .0409834     2.97   0.003     .0414615    .2022201
------------------------------------------------------------------------------

. 
. combomarginsplot "file20D" "file19D" "file18D" "file17D" "file16D", labels("Donate" "Volunteer" "Display sign" "Attend rally" "Persuade others") recast(scatter
> ) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("Marginal Effect of Pride by Mode among Dems") title("") legend(pos(6) row(1)) 
> xsize(8) ysize(6) xlabel(-.6(.2).6) plot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) 
> ci2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(pride16Dem)

  Variables that uniquely identify margins: web _filenumber

. 
. svy: logit persuade c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      8.50
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.435456   .4104303     3.50   0.000     .6304245    2.240488
        1.web |   .6689749   .2473547     2.70   0.007     .1838051    1.154145
              |
  web#c.proud |
           1  |  -.5767897   .4820625    -1.20   0.232    -1.522323     .368744
              |
  campaignint |   1.089415   .2647793     4.11   0.000     .5700677    1.608762
       polint |   .5204814   .3543072     1.47   0.142    -.1744687    1.215432
      polknow |   .3545316   .2616973     1.35   0.176    -.1587704    .8678336
       gender |  -.1807914   .1407664    -1.28   0.199    -.4568954    .0953126
        White |   .1344072    .291153     0.46   0.644    -.4366701    .7054844
       latinx |   .1935115   .4212028     0.46   0.646    -.6326501    1.019673
         ageN |   .2183228   .4442107     0.49   0.623    -.6529672    1.089613
    education |  -.0549582    .575541    -0.10   0.924    -1.183844    1.073928
income_norm01 |  -.0011341   .2683796    -0.00   0.997     -.527543    .5252748
        _cons |  -2.274077   .5393439    -4.22   0.000    -3.331964   -1.216189
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file16R")

Average marginal effects

Number of strata =     1                           Number of obs   =     1,578
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Subpop. no. obs =     1,367
                                                   Subpop. size    =         .
Model VCE: Linearized                              Design df       =     1,615

Expression: Pr(persuade), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |    .310966   .0811413     3.83   0.000     .1518127    .4701193
          2  |   .1933946   .0591885     3.27   0.001     .0773002     .309489
------------------------------------------------------------------------------

. svy: logit rally c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      2.25
                                                   Prob > F        =    0.0082

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.118906   .6797623     1.65   0.100    -.2144032    2.452215
        1.web |   .2178037     .52042     0.42   0.676    -.8029659    1.238573
              |
  web#c.proud |
           1  |  -.0364718   .7999974    -0.05   0.964    -1.605614     1.53267
              |
  campaignint |   .4415638   .5077533     0.87   0.385    -.5543607    1.437488
       polint |   .4049589   .7105658     0.57   0.569    -.9887689    1.798687
      polknow |  -.1420676   .5471596    -0.26   0.795    -1.215285    .9311498
       gender |   .0608763   .2760471     0.22   0.825    -.4805717    .6023244
        White |  -.4574348   .5086859    -0.90   0.369    -1.455189     .540319
       latinx |  -.4664814   .8414519    -0.55   0.579    -2.116934    1.183971
         ageN |  -1.703576   .8813739    -1.93   0.053    -3.432332    .0251812
    education |   1.110309   .9717493     1.14   0.253    -.7957129    3.016331
income_norm01 |   .9445974   .6449425     1.46   0.143    -.3204148     2.20961
        _cons |    -4.1658   1.230556    -3.39   0.001    -6.579454   -1.752145
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file17R")

Average marginal effects

Number of strata =     1                           Number of obs   =     1,578
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Subpop. no. obs =     1,367
                                                   Subpop. size    =         .
Model VCE: Linearized                              Design df       =     1,615

Expression: Pr(rally), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .0509625   .0363405     1.40   0.161    -.0203169     .122242
          2  |   .0585764   .0263953     2.22   0.027     .0068039     .110349
------------------------------------------------------------------------------

. svy: logit button c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      5.34
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   2.043422   .6446126     3.17   0.002     .7790574    3.307787
        1.web |      .5949   .4914719     1.21   0.226    -.3690897     1.55889
              |
  web#c.proud |
           1  |  -.2202534   .8035801    -0.27   0.784    -1.796423    1.355916
              |
  campaignint |   .0586653   .4473535     0.13   0.896    -.8187891    .9361197
       polint |   1.112354   .5785055     1.92   0.055     -.022346    2.247055
      polknow |   .2635325    .372169     0.71   0.479    -.4664523    .9935173
       gender |  -.1379658   .2129198    -0.65   0.517     -.555594    .2796623
        White |   .4628346   .4461563     1.04   0.300    -.4122715    1.337941
       latinx |   .7421812   .5850172     1.27   0.205    -.4052914    1.889654
         ageN |  -1.724509   .6829561    -2.53   0.012    -3.064082   -.3849355
    education |    -.68319   .7169918    -0.95   0.341    -2.089522    .7231421
income_norm01 |  -1.049345    .380411    -2.76   0.006    -1.795496   -.3031938
        _cons |  -2.833774   .7451216    -3.80   0.000     -4.29528   -1.372267
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file18R")

Average marginal effects

Number of strata =     1                           Number of obs   =     1,578
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Subpop. no. obs =     1,367
                                                   Subpop. size    =         .
Model VCE: Linearized                              Design df       =     1,615

Expression: Pr(button), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |    .151561   .0589549     2.57   0.010     .0359249    .2671972
          2  |   .1883421   .0423816     4.44   0.000     .1052135    .2714708
------------------------------------------------------------------------------

. svy: logit volunteer c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      1.24
                                                   Prob > F        =    0.2523

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .6342418   .7151073     0.89   0.375    -.7683938    2.036878
        1.web |   .6196687   .7306552     0.85   0.397    -.8134633    2.052801
              |
  web#c.proud |
           1  |  -.7632938   1.186978    -0.64   0.520    -3.091473    1.564885
              |
  campaignint |   .7131684   .7018031     1.02   0.310    -.6633721    2.089709
       polint |  -.1006194   1.023636    -0.10   0.922    -2.108414    1.907175
      polknow |    .604514   .6559796     0.92   0.357    -.6821468    1.891175
       gender |  -.2285608   .3974637    -0.58   0.565     -1.00816     .551038
        White |   .3594577   .7471342     0.48   0.630    -1.105997    1.824912
       latinx |   .0965286   1.163896     0.08   0.934    -2.186377    2.379434
         ageN |  -1.798535   1.359459    -1.32   0.186    -4.465024    .8679537
    education |  -.2865663   1.513736    -0.19   0.850     -3.25566    2.682528
income_norm01 |  -.1361659   .9608636    -0.14   0.887    -2.020837    1.748505
        _cons |  -4.006839   1.738423    -2.30   0.021    -7.416642   -.5970368
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file19R")

Average marginal effects

Number of strata =     1                           Number of obs   =     1,578
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Subpop. no. obs =     1,367
                                                   Subpop. size    =         .
Model VCE: Linearized                              Design df       =     1,615

Expression: Pr(volunteer), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |   .0130341   .0174997     0.74   0.456    -.0212903    .0473586
          2  |  -.0034512   .0232535    -0.15   0.882    -.0490614    .0421591
------------------------------------------------------------------------------

. svy: logit donatecand c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,334.0214
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       4.17
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |    2.35351   .6478372     3.63   0.000      1.08282    3.624201
        1.web |   .3990849   .5043162     0.79   0.429    -.5900985    1.388268
              |
  web#c.proud |
           1  |  -.6424613   .7966308    -0.81   0.420    -2.205001    .9200781
              |
  campaignint |   1.596381   .6250526     2.55   0.011     .3703811    2.822381
       polint |  -.2231292   .5475218    -0.41   0.684    -1.297057    .8507992
      polknow |   1.092613   .3638804     3.00   0.003      .378885     1.80634
       gender |   .0391974   .2186992     0.18   0.858    -.3897667    .4681616
        White |  -.6805162   .3723503    -1.83   0.068    -1.410857    .0498246
       latinx |   .2017753   .5389388     0.37   0.708     -.855318    1.258869
         ageN |   3.660744   .9274979     3.95   0.000     1.841517     5.47997
    education |   2.162234   .7644354     2.83   0.005     .6628433    3.661624
income_norm01 |   .0305121   .4876194     0.06   0.950    -.9259216    .9869457
        _cons |  -8.147327   1.377482    -5.91   0.000    -10.84917   -5.445485
-------------------------------------------------------------------------------

. margins, dydx(proud) at(web=(0 1)) saving("file20R")

Average marginal effects

Number of strata =     1                          Number of obs   =      1,577
Number of PSUs   = 1,615                          Population size = 1,334.0214
                                                  Subpop. no. obs =      1,366
                                                  Subpop. size    =          .
Model VCE: Linearized                             Design df       =      1,614

Expression: Pr(donatecand), predict()
dy/dx wrt:  proud
1._at: web = 0
2._at: web = 1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
proud        |
         _at |
          1  |    .177632   .0552841     3.21   0.001     .0691958    .2860681
          2  |   .1352888   .0326123     4.15   0.000     .0713219    .1992558
------------------------------------------------------------------------------

. 
. combomarginsplot "file20R" "file19R" "file18R" "file17R" "file16R", labels("Donate" "Volunteer" "Display sign" "Attend rally" "Persuade others") recast(scatter
> ) x(_filenumber) horizontal scheme(plottig) ytitle("") xline(0, lc(red)) xtitle("Marginal Effect of Pride by Mode among Reps") title("") legend(pos(6) row(1)) 
> xsize(8) ysize(6) xlabel(-.6(.2).6) plot1opts(mcolor(white) mlcolor(black) msymbol(D)) ci1opts(msize(tiny)) plot2opts(mcolor(black) mlcolor(black) msymbol(D)) 
> ci2opts(msize(tiny) col(black)) offset(0.25) xline(-1 1, lc(white) lp(solid)) name(pride16Rep)

  Variables that uniquely identify margins: web _filenumber

. 
. 
. grc1leg anger16Dem anger16Rep fear16Dem fear16Rep hope16Dem hope16Rep pride16Dem pride16Rep, scheme(plottig) xcommon ycommon col(2) imargin(small)

. 
. 
. *********************************APPENDIX****************************************

. 
. *********************

. ** Table B3 -- 2016 **

. *********************

. 
. svyset [pweight=V160101]

Sampling weights: V160101
             VCE: linearized
     Single unit: missing
        Strata 1: <one>
 Sampling unit 1: <observations>
           FPC 1: <zero>

. 
. svy: reg anger web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                           Number of obs   =     1,791
Number of PSUs   = 1,791                           Population size = 1,829.147
                                                   Design df       =     1,790
                                                   F(10, 1781)     =     15.20
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1026

-------------------------------------------------------------------------------
              |             Linearized
        anger | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .1002454   .0198505     5.05   0.000     .0613129    .1391779
  campaignint |   .1180734   .0341675     3.46   0.001      .051061    .1850859
       polint |   .0673289   .0425247     1.58   0.114    -.0160744    .1507321
      polknow |   .0458501   .0345056     1.33   0.184    -.0218254    .1135256
       gender |   .0799393   .0179888     4.44   0.000      .044658    .1152206
        White |  -.1022219   .0212746    -4.80   0.000    -.1439475   -.0604963
       latinx |   .0364742   .0273761     1.33   0.183    -.0172183    .0901666
         ageN |  -.1145268   .0507102    -2.26   0.024    -.2139841   -.0150695
    education |   .1095012   .0597143     1.83   0.067    -.0076159    .2266183
income_norm01 |   .0908028   .0349848     2.60   0.010     .0221875     .159418
        _cons |   .4258539   .0505375     8.43   0.000     .3267352    .5249726
-------------------------------------------------------------------------------

. outreg2 using tabD16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Anger toward Out-Party Candidate") append
tabD16.doc
dir : seeout

. svy: reg fear web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,788
Number of PSUs   = 1,788                          Population size = 1,828.4554
                                                  Design df       =      1,787
                                                  F(10, 1778)     =      11.50
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0841

-------------------------------------------------------------------------------
              |             Linearized
         fear | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .1172374   .0216659     5.41   0.000     .0747443    .1597305
  campaignint |   .1304352    .037995     3.43   0.001      .055916    .2049544
       polint |   .0157443   .0483002     0.33   0.744    -.0789865    .1104751
      polknow |   .0240438    .036862     0.65   0.514    -.0482534     .096341
       gender |   .0620864   .0199257     3.12   0.002     .0230063    .1011666
        White |  -.0917422    .022868    -4.01   0.000     -.136593   -.0468914
       latinx |  -.0014586   .0338517    -0.04   0.966    -.0678517    .0649345
         ageN |   .0701162   .0574481     1.22   0.222    -.0425564    .1827888
    education |   .1990571   .0640239     3.11   0.002     .0734875    .3246267
income_norm01 |   .0783163   .0376784     2.08   0.038     .0044178    .1522147
        _cons |   .2881705   .0551462     5.23   0.000     .1800127    .3963283
-------------------------------------------------------------------------------

. outreg2 using tabD16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Fear toward Out-Party Candidate") append
tabD16.doc
dir : seeout

. svy: reg hopeful web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                           Number of obs   =     1,790
Number of PSUs   = 1,790                           Population size = 1,829.837
                                                   Design df       =     1,789
                                                   F(10, 1780)     =     13.17
                                                   Prob > F        =    0.0000
                                                   R-squared       =    0.1047

-------------------------------------------------------------------------------
              |             Linearized
      hopeful | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0151073   .0186831     0.81   0.419    -.0215357    .0517503
  campaignint |   .2071938   .0350935     5.90   0.000     .1383651    .2760224
       polint |   .0343754   .0445174     0.77   0.440    -.0529361     .121687
      polknow |   .0470729   .0328034     1.43   0.151    -.0172642    .1114099
       gender |   .0326876   .0184603     1.77   0.077    -.0035184    .0688937
        White |  -.1073595   .0219096    -4.90   0.000    -.1503305   -.0643885
       latinx |   .0025336   .0313089     0.08   0.936    -.0588723    .0639395
         ageN |    .179551   .0537775     3.34   0.001     .0740776    .2850243
    education |   .0222804   .0600718     0.37   0.711     -.095538    .1400987
income_norm01 |   .0130548   .0345412     0.38   0.706    -.0546906    .0808002
        _cons |   .2569812   .0512017     5.02   0.000     .1565598    .3574027
-------------------------------------------------------------------------------

. outreg2 using tabD16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Hope toward In-Party Candidate") append
tabD16.doc
dir : seeout

. svy: reg proud web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,792
Number of PSUs   = 1,792                          Population size = 1,830.7972
                                                  Design df       =      1,791
                                                  F(10, 1782)     =      17.01
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.1345

-------------------------------------------------------------------------------
              |             Linearized
        proud | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   -.006734   .0192136    -0.35   0.726    -.0444174    .0309495
  campaignint |   .2195434   .0356995     6.15   0.000     .1495262    .2895605
       polint |   .0520017   .0460624     1.13   0.259      -.03834    .1423434
      polknow |   .0903791   .0334713     2.70   0.007     .0247321     .156026
       gender |   .0513276   .0183766     2.79   0.005     .0152857    .0873694
        White |  -.1440548   .0219135    -6.57   0.000    -.1870334   -.1010761
       latinx |  -.0125866   .0305611    -0.41   0.680    -.0725258    .0473526
         ageN |   .1754688   .0536617     3.27   0.001     .0702226     .280715
    education |    .003659   .0616692     0.06   0.953    -.1172922    .1246101
income_norm01 |  -.0259128   .0359619    -0.72   0.471    -.0964445     .044619
        _cons |   .2562704   .0514707     4.98   0.000     .1553214    .3572193
-------------------------------------------------------------------------------

. outreg2 using tabD16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Pride toward In-Party Candidate") append
tabD16.doc
dir : seeout

. 
. *********************

. ** Table B4 -- 2016 **

. *********************

. 
. svy: reg anger web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,576
Number of PSUs   = 1,576                          Population size = 1,538.0834
                                                  Design df       =      1,575
                                                  F(10, 1566)     =       7.78
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0751

-------------------------------------------------------------------------------
              |             Linearized
        anger | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0643238   .0211233     3.05   0.002      .022891    .1057566
  campaignint |   .0696518   .0359015     1.94   0.053    -.0007681    .1400716
       polint |   .1645164   .0481131     3.42   0.001     .0701439     .258889
      polknow |  -.0128353   .0357875    -0.36   0.720    -.0830314    .0573608
       gender |   .0222158   .0189673     1.17   0.242     -.014988    .0594197
        White |   .1401492   .0420321     3.33   0.001     .0577045    .2225939
       latinx |  -.0480113    .059857    -0.80   0.423    -.1654192    .0693965
         ageN |  -.1162497     .05657    -2.05   0.040    -.2272101   -.0052893
    education |  -.2037977   .0688592    -2.96   0.003    -.3388631   -.0687322
income_norm01 |   .0115735   .0372549     0.31   0.756    -.0615009    .0846478
        _cons |   .5221726   .0665867     7.84   0.000     .3915647    .6527805
-------------------------------------------------------------------------------

. outreg2 using tabR16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Anger toward Out-Party Candidate") append
tabR16.doc
dir : seeout

. svy: reg fear web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,579
Number of PSUs   = 1,579                          Population size = 1,539.3775
                                                  Design df       =      1,578
                                                  F(10, 1569)     =       8.43
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0686

-------------------------------------------------------------------------------
              |             Linearized
         fear | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0603229   .0224571     2.69   0.007     .0162741    .1043718
  campaignint |   .0654347   .0377978     1.73   0.084    -.0087045    .1395739
       polint |   .1965942   .0478007     4.11   0.000     .1028345    .2903538
      polknow |  -.0242201    .037575    -0.64   0.519    -.0979224    .0494821
       gender |   .0281047   .0205994     1.36   0.173    -.0123004    .0685097
        White |   .1101438   .0448574     2.46   0.014     .0221574    .1981302
       latinx |   -.044474    .060102    -0.74   0.459    -.1623623    .0734142
         ageN |  -.0030536   .0621142    -0.05   0.961    -.1248887    .1187815
    education |  -.2422674   .0717565    -3.38   0.001    -.3830155   -.1015192
income_norm01 |  -.0357518     .03853    -0.93   0.354    -.1113271    .0398236
        _cons |   .4853919   .0724703     6.70   0.000     .3432438      .62754
-------------------------------------------------------------------------------

. outreg2 using tabR16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Fear toward Out-Party Candidate") append
tabR16.doc
dir : seeout

. svy: reg hopeful web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,578                          Population size = 1,538.8918
                                                  Design df       =      1,577
                                                  F(10, 1568)     =      21.17
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.1433

-------------------------------------------------------------------------------
              |             Linearized
      hopeful | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0636713   .0203234     3.13   0.002     .0238076     .103535
  campaignint |   .1956429   .0359124     5.45   0.000     .1252018     .266084
       polint |   .1507674   .0466255     3.23   0.001      .059313    .2422218
      polknow |  -.0837815   .0329935    -2.54   0.011    -.1484971   -.0190659
       gender |  -.0216152   .0187908    -1.15   0.250    -.0584727    .0152423
        White |   .0917623   .0381754     2.40   0.016     .0168824    .1666423
       latinx |  -.0362831   .0565441    -0.64   0.521    -.1471926    .0746264
         ageN |   .0650697   .0569422     1.14   0.253    -.0466207    .1767601
    education |  -.3512986   .0680686    -5.16   0.000    -.4848131   -.2177842
income_norm01 |  -.0601919   .0349536    -1.72   0.085    -.1287523    .0083686
        _cons |   .4331723   .0678643     6.38   0.000     .3000584    .5662861
-------------------------------------------------------------------------------

. outreg2 using tabR16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Hope toward In-Party Candidate") append
tabR16.doc
dir : seeout

. svy: reg proud web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,578                          Population size = 1,537.8061
                                                  Design df       =      1,577
                                                  F(10, 1568)     =      19.33
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.1299

-------------------------------------------------------------------------------
              |             Linearized
        proud | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0818275   .0195595     4.18   0.000     .0434622    .1201928
  campaignint |   .1796924   .0355623     5.05   0.000     .1099379    .2494468
       polint |   .1294518   .0473448     2.73   0.006     .0365865    .2223172
      polknow |   -.114106   .0339039    -3.37   0.001    -.1806076   -.0476045
       gender |  -.0177048   .0184858    -0.96   0.338    -.0539642    .0185545
        White |   .0808625   .0365035     2.22   0.027      .009262    .1524629
       latinx |   -.041043   .0541088    -0.76   0.448    -.1471757    .0650898
         ageN |   .0731022   .0593749     1.23   0.218    -.0433599    .1895643
    education |    -.33945   .0692129    -4.90   0.000    -.4752089   -.2036911
income_norm01 |  -.0435116   .0352963    -1.23   0.218    -.1127441     .025721
        _cons |    .363234   .0648724     5.60   0.000     .2359889    .4904792
-------------------------------------------------------------------------------

. outreg2 using tabR16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) title("DV: Pride toward In-Party Candidate") append
tabR16.doc
dir : seeout

. 
. *********************

. ** Table B13 -- 2016 **

. *********************

. 
. svyset [pweight=V160102]

Sampling weights: V160102
             VCE: linearized
     Single unit: missing
        Strata 1: <one>
 Sampling unit 1: <observations>
           FPC 1: <zero>

. 
. svy: logit persuade c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,599.9581
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       9.34
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .5343922   .3702504     1.44   0.149    -.1917663    1.260551
        1.web |  -.1002641   .3221981    -0.31   0.756    -.7321793    .5316512
              |
  web#c.anger |
           1  |   .1178166   .4431134     0.27   0.790    -.7512456    .9868787
              |
  campaignint |   1.354049   .2647129     5.12   0.000     .8348776    1.873221
       polint |    .592086   .3296837     1.80   0.073    -.0545106    1.238683
      polknow |   .7146296   .2570783     2.78   0.005     .2104313    1.218828
       gender |   .1022145   .1401692     0.73   0.466    -.1726942    .3771232
        White |    .277535   .1586649     1.75   0.080    -.0336487    .5887187
       latinx |   .0750466   .2339252     0.32   0.748    -.3837422    .5338354
         ageN |  -1.094857   .4150844    -2.64   0.008    -1.908947   -.2807677
    education |   .4571222   .4916381     0.93   0.353    -.5071097    1.421354
income_norm01 |   .0479174   .2593299     0.18   0.853    -.4606969    .5565316
        _cons |  -1.974852   .4282101    -4.61   0.000    -2.814685   -1.135019
-------------------------------------------------------------------------------

. outreg2 using tabDemanger16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemanger16.doc
dir : seeout

. svy: logit rally c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,827
Number of PSUs   = 1,827                          Population size = 1,598.9989
                                                  Design df       =      1,826
                                                  F(12, 1815)     =       3.56
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |  -.7601046   .5765898    -1.32   0.188    -1.890949    .3707403
        1.web |  -.9586753   .5945461    -1.61   0.107    -2.124737    .2073865
              |
  web#c.anger |
           1  |   1.184198   .7657217     1.55   0.122    -.3175842    2.685981
              |
  campaignint |   1.503391   .5756642     2.61   0.009     .3743619    2.632421
       polint |    .925496   .7226669     1.28   0.200    -.4918446    2.342836
      polknow |    .078918   .4083663     0.19   0.847    -.7219962    .8798321
       gender |   .0435489   .2145492     0.20   0.839    -.3772387    .4643366
        White |   .1849436     .28522     0.65   0.517    -.3744482    .7443355
       latinx |  -.0191288   .4312576    -0.04   0.965    -.8649388    .8266812
         ageN |  -2.207567   .7116209    -3.10   0.002    -3.603243   -.8118905
    education |   .8136993   .8758531     0.93   0.353    -.9040798    2.531479
income_norm01 |  -.3962659   .4215209    -0.94   0.347     -1.22298    .4304479
        _cons |  -3.015866    .755731    -3.99   0.000    -4.498054   -1.533678
-------------------------------------------------------------------------------

. outreg2 using tabDemanger16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemanger16.doc
dir : seeout

. svy: logit button c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,599.9581
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       4.85
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .7033358   .6524924     1.08   0.281    -.5763736    1.983045
        1.web |   .0425708   .6147833     0.07   0.945    -1.163181    1.248323
              |
  web#c.anger |
           1  |   .3001683   .7748193     0.39   0.699    -1.219456    1.819793
              |
  campaignint |   1.398399   .3809456     3.67   0.000     .6512644    2.145534
       polint |   .8455602   .4922229     1.72   0.086    -.1198184    1.810939
      polknow |   .6251417   .3214947     1.94   0.052    -.0053942    1.255678
       gender |   .0388982   .1968609     0.20   0.843    -.3471979    .4249943
        White |   -.057111    .222117    -0.26   0.797    -.4927409    .3785189
       latinx |  -.4508434   .3436542    -1.31   0.190     -1.12484    .2231529
         ageN |  -1.411493    .581418    -2.43   0.015    -2.551807   -.2711796
    education |  -.6897887   .6352446    -1.09   0.278    -1.935671    .5560933
income_norm01 |  -.1364993   .3362699    -0.41   0.685    -.7960131    .5230146
        _cons |  -3.225653   .7357481    -4.38   0.000    -4.668649   -1.782657
-------------------------------------------------------------------------------

. outreg2 using tabDemanger16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemanger16.doc
dir : seeout

. svy: logit volunteer c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,599.9581
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       3.93
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.026981   1.136845     0.90   0.366    -1.202672    3.256633
        1.web |   .8148207   1.109301     0.73   0.463    -1.360811    2.990452
              |
  web#c.anger |
           1  |  -.8229037   1.288628    -0.64   0.523    -3.350242    1.704435
              |
  campaignint |   1.788112   .8384941     2.13   0.033     .1436045     3.43262
       polint |   1.048917   1.015529     1.03   0.302    -.9428027    3.040637
      polknow |   .0277809   .5919877     0.05   0.963    -1.133263    1.188825
       gender |  -.1509171   .3149821    -0.48   0.632    -.7686799    .4668456
        White |  -.6334908   .3341754    -1.90   0.058    -1.288897    .0219152
       latinx |  -.3871859   .5541162    -0.70   0.485    -1.473954    .6995818
         ageN |  -.9807477   .9266704    -1.06   0.290    -2.798192     .836697
    education |   2.609106   1.143884     2.28   0.023     .3656486    4.852562
income_norm01 |  -.3938606   .5119075    -0.77   0.442    -1.397846    .6101247
        _cons |  -6.943333   1.467367    -4.73   0.000    -9.821227    -4.06544
-------------------------------------------------------------------------------

. outreg2 using tabDemanger16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemanger16.doc
dir : seeout

. svy: logit donatecand c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,826
Number of PSUs   = 1,826                          Population size = 1,598.1421
                                                  Design df       =      1,825
                                                  F(12, 1814)     =       6.94
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .9355193   .4958923     1.89   0.059    -.0370567    1.908095
        1.web |  -.4141821   .5658933    -0.73   0.464    -1.524049    .6956844
              |
  web#c.anger |
           1  |  -.0593617   .6823645    -0.09   0.931    -1.397659    1.278936
              |
  campaignint |   .9883632   .6421246     1.54   0.124     -.271013    2.247739
       polint |   1.275356   .6122253     2.08   0.037     .0746196    2.476092
      polknow |   1.732565   .4209387     4.12   0.000     .9069924    2.558137
       gender |  -.1328539   .1787536    -0.74   0.457    -.4834371    .2177292
        White |  -.0676015   .2536292    -0.27   0.790    -.5650355    .4298324
       latinx |  -.4243581   .3762444    -1.13   0.260    -1.162273    .3135567
         ageN |   .4787991    .681584     0.70   0.482    -.8579675    1.815566
    education |   .7228511    .887806     0.81   0.416    -1.018371    2.464074
income_norm01 |    .439269   .4084298     1.08   0.282      -.36177    1.240308
        _cons |  -5.615812   .8636241    -6.50   0.000    -7.309608   -3.922017
-------------------------------------------------------------------------------

. outreg2 using tabDemanger16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemanger16.doc
dir : seeout

. 
. *********************

. ** Table B14 -- 2016 **

. *********************

. 
. svy: logit persuade c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,613
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Design df       =      1,612
                                                  F(12, 1601)     =       7.75
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   1.170308   .3841668     3.05   0.002     .4167895    1.923827
        1.web |   1.203244   .3335697     3.61   0.000     .5489678     1.85752
              |
  web#c.anger |
           1  |  -1.102719   .4649879    -2.37   0.018    -2.014764    -.190675
              |
  campaignint |   1.207343    .261709     4.61   0.000     .6940179    1.720669
       polint |   .5916139    .344933     1.72   0.087    -.0849504    1.268178
      polknow |   .2762146   .2547577     1.08   0.278    -.2234764    .7759057
       gender |  -.2123092   .1390127    -1.53   0.127    -.4849738    .0603554
        White |   .2122185    .271647     0.78   0.435    -.3205999    .7450369
       latinx |   .2325076   .3975542     0.58   0.559    -.5472698    1.012285
         ageN |   .3617656   .4321345     0.84   0.403     -.485839     1.20937
    education |  -.3122208   .5595363    -0.56   0.577    -1.409716    .7852743
income_norm01 |  -.0884856    .264367    -0.33   0.738    -.6070248    .4300536
        _cons |  -2.468228   .5225004    -4.72   0.000     -3.49308   -1.443377
-------------------------------------------------------------------------------

. outreg2 using tabRepanger16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepanger16.doc
dir : seeout

. svy: logit rally c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,613
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Design df       =      1,612
                                                  F(12, 1601)     =       1.57
                                                  Prob > F        =     0.0942

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .5906305   .7834546     0.75   0.451    -.9460661    2.127327
        1.web |   .5566605   .7881005     0.71   0.480    -.9891487     2.10247
              |
  web#c.anger |
           1  |  -.4013125   .9754002    -0.41   0.681    -2.314498    1.511873
              |
  campaignint |   .6122251   .4996289     1.23   0.221    -.3677654    1.592216
       polint |   .5289772   .7362157     0.72   0.473    -.9150632    1.973018
      polknow |  -.2550917   .5406461    -0.47   0.637    -1.315535    .8053514
       gender |   .0419932   .2774446     0.15   0.880    -.5021969    .5861832
        White |  -.4074962   .5101979    -0.80   0.425    -1.408217    .5932246
       latinx |  -.4299135   .8245133    -0.52   0.602    -2.047144    1.187317
         ageN |  -1.582612    .883335    -1.79   0.073    -3.315218    .1499935
    education |   .6902233    .945015     0.73   0.465    -1.163364     2.54381
income_norm01 |     .86706   .6384407     1.36   0.175     -.385201    2.119321
        _cons |  -4.020338   1.264914    -3.18   0.002    -6.501386   -1.539289
-------------------------------------------------------------------------------

. outreg2 using tabRepanger16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepanger16.doc
dir : seeout

. svy: logit button c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,613
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Design df       =      1,612
                                                  F(12, 1601)     =       4.27
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .7856771    .587369     1.34   0.181    -.3664101    1.937764
        1.web |   .1904753   .5716522     0.33   0.739    -.9307842    1.311735
              |
  web#c.anger |
           1  |   .5369527   .7176693     0.75   0.454    -.8707102    1.944616
              |
  campaignint |   .3022351   .4666523     0.65   0.517    -.6130738    1.217544
       polint |   1.273814   .6038898     2.11   0.035     .0893225    2.458306
      polknow |   .1210478   .3695833     0.33   0.743    -.6038665     .845962
       gender |  -.1691369   .2118556    -0.80   0.425    -.5846783    .2464045
        White |   .4373596   .4293709     1.02   0.309    -.4048241    1.279543
       latinx |   .7609857   .5669352     1.34   0.180    -.3510218    1.872993
         ageN |  -1.560275   .6631451    -2.35   0.019    -2.860992   -.2595574
    education |  -1.293142   .7094509    -1.82   0.069    -2.684685    .0984014
income_norm01 |  -1.181941    .383586    -3.08   0.002    -1.934321   -.4295612
        _cons |  -2.256299   .7573623    -2.98   0.003    -3.741818   -.7707809
-------------------------------------------------------------------------------

. outreg2 using tabRepanger16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepanger16.doc
dir : seeout

. svy: logit volunteer c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,613
Number of PSUs   = 1,613                          Population size = 1,333.1409
                                                  Design df       =      1,612
                                                  F(12, 1601)     =       0.97
                                                  Prob > F        =     0.4783

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   .6667652   .9788358     0.68   0.496    -1.253159     2.58669
        1.web |   .5176289   .9078005     0.57   0.569    -1.262964    2.298222
              |
  web#c.anger |
           1  |  -.3487086    1.20726    -0.29   0.773    -2.716672    2.019255
              |
  campaignint |   .6802386    .769698     0.88   0.377    -.8294753    2.189953
       polint |  -.1463822   1.070624    -0.14   0.891    -2.246344     1.95358
      polknow |   .6052565   .6694516     0.90   0.366    -.7078304    1.918343
       gender |  -.2378504   .3960012    -0.60   0.548    -1.014582     .538881
        White |   .3247211   .7284349     0.45   0.656    -1.104058      1.7535
       latinx |   .1389287   1.165534     0.12   0.905    -2.147192     2.42505
         ageN |  -1.722597   1.330235    -1.29   0.196     -4.33177    .8865746
    education |  -.1899583   1.445321    -0.13   0.895    -3.024863    2.644947
income_norm01 |  -.1362162   .9114509    -0.15   0.881    -1.923969    1.651537
        _cons |  -4.199323   1.806332    -2.32   0.020    -7.742328    -.656318
-------------------------------------------------------------------------------

. outreg2 using tabRepanger16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepanger16.doc
dir : seeout

. svy: logit donatecand c.anger##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,612
Number of PSUs   = 1,612                          Population size = 1,332.7643
                                                  Design df       =      1,611
                                                  F(12, 1600)     =       5.46
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        anger |   2.699537   .8157109     3.31   0.001     1.099571    4.299504
        1.web |    .860183   .8009987     1.07   0.283    -.7109259    2.431292
              |
  web#c.anger |
           1  |  -1.015479   .9512415    -1.07   0.286     -2.88128    .8503217
              |
  campaignint |     1.8651   .6369025     2.93   0.003     .6158554    3.114345
       polint |  -.3216029   .5722263    -0.56   0.574    -1.443989    .8007833
      polknow |   .9106442   .3738442     2.44   0.015     .1773722    1.643916
       gender |   .0066229   .2161572     0.03   0.976    -.4173558    .4306017
        White |  -.7602173   .3815185    -1.99   0.046    -1.508542   -.0118925
       latinx |   .2630985   .5447382     0.48   0.629    -.8053715    1.331568
         ageN |   3.730941   .9001559     4.14   0.000     1.965341    5.496541
    education |    1.66854   .7555538     2.21   0.027     .1865684    3.150512
income_norm01 |  -.0379571   .4868347    -0.08   0.938    -.9928529    .9169388
        _cons |  -8.616266   1.416463    -6.08   0.000    -11.39457   -5.837962
-------------------------------------------------------------------------------

. outreg2 using tabRepanger16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepanger16.doc
dir : seeout

. 
. *********************

. ** Table B15 -- 2016 **

. *********************

. 
. svy: logit persuade c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,826
Number of PSUs   = 1,826                          Population size = 1,600.1414
                                                  Design df       =      1,825
                                                  F(12, 1814)     =       9.62
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .1628709    .343895     0.47   0.636    -.5115983      .83734
        1.web |  -.3710744   .3059882    -1.21   0.225    -.9711982    .2290494
              |
   web#c.fear |
           1  |   .5664959   .4165139     1.36   0.174    -.2503981     1.38339
              |
  campaignint |   1.386682   .2667208     5.20   0.000     .8635724    1.909792
       polint |   .5986223   .3305739     1.81   0.070    -.0497205    1.246965
      polknow |    .748512   .2585077     2.90   0.004     .2415101    1.255514
       gender |   .1306609   .1386928     0.94   0.346    -.1413524    .4026741
        White |   .2768747   .1579618     1.75   0.080    -.0329301    .5866796
       latinx |   .1205489   .2312038     0.52   0.602     -.332903    .5740007
         ageN |  -1.199648   .4111033    -2.92   0.004    -2.005931    -.393366
    education |   .4418515   .4933743     0.90   0.371    -.5257861    1.409489
income_norm01 |   .0255659   .2590986     0.10   0.921    -.4825951    .5337269
        _cons |  -1.738449    .421223    -4.13   0.000    -2.564579   -.9123194
-------------------------------------------------------------------------------

. outreg2 using tabDemfear16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemfear16.doc
dir : seeout

. svy: logit rally c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,825
Number of PSUs   = 1,825                          Population size = 1,599.1822
                                                  Design df       =      1,824
                                                  F(12, 1813)     =       3.73
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |  -.8024842   .4907494    -1.64   0.102    -1.764974    .1600056
        1.web |  -.8203993   .5191344    -1.58   0.114     -1.83856    .1977609
              |
   web#c.fear |
           1  |   1.068008   .6755188     1.58   0.114    -.2568635     2.39288
              |
  campaignint |   1.554703   .5762071     2.70   0.007     .4246079    2.684798
       polint |    .908809   .7098175     1.28   0.201    -.4833316     2.30095
      polknow |   .1123293   .4165454     0.27   0.787    -.7046268    .9292854
       gender |   .0692554   .2169179     0.32   0.750    -.3561782     .494689
        White |   .1720721   .2879798     0.60   0.550    -.3927328    .7368771
       latinx |    .018659   .4265693     0.04   0.965    -.8179567    .8552746
         ageN |   -2.18737   .7161625    -3.05   0.002    -3.591955   -.7827854
    education |   .8218328   .8850687     0.93   0.353    -.9140218    2.557687
income_norm01 |  -.4204728   .4238552    -0.99   0.321    -1.251765    .4108197
        _cons |  -3.085016   .7326751    -4.21   0.000    -4.521987   -1.648046
-------------------------------------------------------------------------------

. outreg2 using tabDemfear16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemfear16.doc
dir : seeout

. svy: logit button c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,826
Number of PSUs   = 1,826                          Population size = 1,600.1414
                                                  Design df       =      1,825
                                                  F(12, 1814)     =       4.58
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .3393747   .5150632     0.66   0.510    -.6708006     1.34955
        1.web |   .5720679   .4672332     1.22   0.221       -.3443    1.488436
              |
   web#c.fear |
           1  |  -.3508396   .6121926    -0.57   0.567    -1.551511    .8498321
              |
  campaignint |   1.487301   .3861232     3.85   0.000     .7300112    2.244591
       polint |    .908844   .4955284     1.83   0.067    -.0630184    1.880706
      polknow |    .634835   .3159503     2.01   0.045     .0151729    1.254497
       gender |    .088695   .1967292     0.45   0.652     -.297143     .474533
        White |  -.1205644   .2266235    -0.53   0.595     -.565033    .3239041
       latinx |  -.4216757   .3360251    -1.25   0.210     -1.08071    .2373584
         ageN |   -1.51628    .593039    -2.56   0.011    -2.679386    -.353173
    education |  -.5891692   .6376173    -0.92   0.356    -1.839706    .6613671
income_norm01 |   -.071978   .3328138    -0.22   0.829     -.724714    .5807581
        _cons |  -3.135613   .6952657    -4.51   0.000    -4.499213   -1.772012
-------------------------------------------------------------------------------

. outreg2 using tabDemfear16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemfear16.doc
dir : seeout

. svy: logit volunteer c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,826
Number of PSUs   = 1,826                          Population size = 1,600.1414
                                                  Design df       =      1,825
                                                  F(12, 1814)     =       4.10
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |  -.8918624   .7228314    -1.23   0.217    -2.309526    .5258014
        1.web |  -.0722712   .7171733    -0.10   0.920    -1.478838    1.334296
              |
   web#c.fear |
           1  |   .5242943   .9165081     0.57   0.567    -1.273221    2.321809
              |
  campaignint |   1.965854   .8272733     2.38   0.018     .3433517    3.588356
       polint |   1.093535   1.010435     1.08   0.279    -.8881963    3.075265
      polknow |    .025179   .5915375     0.04   0.966    -1.134983    1.185341
       gender |  -.0691593   .3186991    -0.22   0.828    -.6942126     .555894
        White |  -.6622941   .3366228    -1.97   0.049      -1.3225   -.0020877
       latinx |  -.3549729   .5529294    -0.64   0.521    -1.439414    .7294681
         ageN |  -1.064183   .9033844    -1.18   0.239    -2.835959    .7075928
    education |   2.823805    1.15192     2.45   0.014     .5645858    5.083024
income_norm01 |  -.3660128   .5169193    -0.71   0.479    -1.379828    .6478027
        _cons |  -5.982356   1.279305    -4.68   0.000    -8.491411     -3.4733
-------------------------------------------------------------------------------

. outreg2 using tabDemfear16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemfear16.doc
dir : seeout

. svy: logit donatecand c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,824
Number of PSUs   = 1,824                          Population size = 1,598.3254
                                                  Design df       =      1,823
                                                  F(12, 1812)     =       6.59
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .4758779    .507859     0.94   0.349    -.5201687    1.471924
        1.web |  -.5297482   .5670001    -0.93   0.350    -1.641786    .5822899
              |
   web#c.fear |
           1  |   .1370395   .6725764     0.20   0.839    -1.182062    1.456141
              |
  campaignint |   1.048958   .6529832     1.61   0.108    -.2317159    2.329632
       polint |   1.330565   .6200133     2.15   0.032     .1145542    2.546576
      polknow |   1.752435   .4318842     4.06   0.000     .9053951    2.599475
       gender |  -.1111781   .1801566    -0.62   0.537    -.4645131    .2421568
        White |  -.0664107   .2533894    -0.26   0.793    -.5633748    .4305533
       latinx |  -.3760333   .3755124    -1.00   0.317    -1.112513    .3604464
         ageN |   .3154714   .6829745     0.46   0.644    -1.024023    1.654966
    education |   .7452502   .8786634     0.85   0.396    -.9780425    2.468543
income_norm01 |   .4431423   .4042088     1.10   0.273    -.3496188    1.235903
        _cons |  -5.337958   .9093833    -5.87   0.000      -7.1215   -3.554415
-------------------------------------------------------------------------------

. outreg2 using tabDemfear16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemfear16.doc
dir : seeout

. 
. *********************

. ** Table B16 -- 2016 **

. *********************

. 
. svy: logit persuade c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      8.34
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.440056   .3545849     4.06   0.000     .7445612    2.135551
        1.web |   1.323541   .3004719     4.40   0.000     .7341853    1.912897
              |
   web#c.fear |
           1  |  -1.384068   .4252257    -3.25   0.001     -2.21812   -.5500161
              |
  campaignint |   1.237032   .2655432     4.66   0.000     .7161868    1.757878
       polint |   .5626144   .3495745     1.61   0.108    -.1230529    1.248282
      polknow |   .2976264   .2552475     1.17   0.244    -.2030246    .7982775
       gender |  -.2101915   .1398728    -1.50   0.133    -.4845428    .0641597
        White |   .2144418    .274638     0.78   0.435    -.3242425    .7531262
       latinx |   .2561236   .4016333     0.64   0.524    -.5316537    1.043901
         ageN |   .3066391   .4351804     0.70   0.481    -.5469386    1.160217
    education |  -.3176822   .5564727    -0.57   0.568    -1.409167    .7738023
income_norm01 |  -.0760086   .2648773    -0.29   0.774    -.5955479    .4435306
        _cons |  -2.574688    .522586    -4.93   0.000    -3.599706    -1.54967
-------------------------------------------------------------------------------

. outreg2 using tabRepfear16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepfear16.doc
dir : seeout

. svy: logit rally c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      1.50
                                                   Prob > F        =    0.1165

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .0921361   .9200882     0.10   0.920    -1.712556    1.896828
        1.web |   .2927792   .7748356     0.38   0.706     -1.22701    1.812568
              |
   web#c.fear |
           1  |   .0148811   1.087616     0.01   0.989    -2.118406    2.148168
              |
  campaignint |   .6361482   .4996852     1.27   0.203    -.3439513    1.616248
       polint |   .5490428   .7301141     0.75   0.452    -.8830278    1.981113
      polknow |  -.2567544    .549119    -0.47   0.640    -1.333815    .8203062
       gender |   .0489501    .279662     0.18   0.861    -.4995885    .5974886
        White |  -.3939407   .5050408    -0.78   0.435    -1.384545    .5966635
       latinx |  -.4379307   .8286753    -0.53   0.597    -2.063323    1.187461
         ageN |  -1.599103    .879427    -1.82   0.069    -3.324041    .1258346
    education |   .6304133   .9540976     0.66   0.509    -1.240986    2.501813
income_norm01 |   .8841622   .6296414     1.40   0.160    -.3508378    2.119162
        _cons |  -3.702044   1.415914    -2.61   0.009    -6.479266   -.9248226
-------------------------------------------------------------------------------

. outreg2 using tabRepfear16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepfear16.doc
dir : seeout

. svy: logit button c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      3.61
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   .8738197   .5558649     1.57   0.116    -.2164725    1.964112
        1.web |   .8250435   .5020219     1.64   0.100    -.1596392    1.809726
              |
   web#c.fear |
           1  |  -.3578268   .6663928    -0.54   0.591    -1.664912    .9492586
              |
  campaignint |   .3061883    .459532     0.67   0.505    -.5951534     1.20753
       polint |   1.319697   .6027726     2.19   0.029     .1373981    2.501996
      polknow |   .1041775   .3746267     0.28   0.781    -.6306279     .838983
       gender |  -.1633819   .2114499    -0.77   0.440    -.5781269    .2513632
        White |   .5781665   .4358842     1.33   0.185    -.2767916    1.433125
       latinx |   .7488249   .5581401     1.34   0.180      -.34593     1.84358
         ageN |  -1.670504   .6693307    -2.50   0.013    -2.983352   -.3576559
    education |  -1.292187     .71894    -1.80   0.072     -2.70234    .1179664
income_norm01 |  -1.111953    .380286    -2.92   0.004    -1.857859   -.3660468
        _cons |   -2.39417   .7607274    -3.15   0.002    -3.886287   -.9020539
-------------------------------------------------------------------------------

. outreg2 using tabRepfear16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepfear16.doc
dir : seeout

. svy: logit volunteer c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      1.29
                                                   Prob > F        =    0.2181

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.144407   .7823453     1.46   0.144    -.3901112    2.678926
        1.web |   .6044782   .8776678     0.69   0.491    -1.117009    2.325966
              |
   web#c.fear |
           1  |  -.5107372   1.026816    -0.50   0.619     -2.52477    1.503295
              |
  campaignint |   .6723823   .7691365     0.87   0.382    -.8362283    2.180993
       polint |  -.2400205   1.099079    -0.22   0.827    -2.395791    1.915751
      polknow |   .6318279   .6762711     0.93   0.350    -.6946332    1.958289
       gender |  -.2432261    .396362    -0.61   0.540    -1.020664    .5342117
        White |   .2927965    .733259     0.40   0.690    -1.145443    1.731036
       latinx |   .1450015   1.176849     0.12   0.902    -2.163309    2.453312
         ageN |  -1.754872   1.322602    -1.33   0.185    -4.349068    .8393245
    education |  -.1309392   1.432395    -0.09   0.927    -2.940488    2.678609
income_norm01 |  -.0914592   .8905761    -0.10   0.918    -1.838265    1.655347
        _cons |  -4.454852   1.712426    -2.60   0.009    -7.813662   -1.096042
-------------------------------------------------------------------------------

. outreg2 using tabRepfear16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepfear16.doc
dir : seeout

. svy: logit donatecand c.fear##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,334.0214
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       4.63
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
         fear |   1.573378   .7296592     2.16   0.031      .142199    3.004557
        1.web |   .3062433   .6841214     0.45   0.654    -1.035616    1.648103
              |
   web#c.fear |
           1  |  -.3214345   .8358451    -0.38   0.701     -1.96089    1.318021
              |
  campaignint |   1.841326   .6240515     2.95   0.003     .6172898    3.065363
       polint |  -.2608015    .574542    -0.45   0.650    -1.387728    .8661253
      polknow |   .9505059   .3780234     2.51   0.012     .2090376    1.691974
       gender |   .0061179    .216447     0.03   0.977    -.4184288    .4306646
        White |  -.6942795   .3687406    -1.88   0.060     -1.41754    .0289812
       latinx |   .2082203   .5432698     0.38   0.702    -.8573681    1.273809
         ageN |   3.654431   .9004476     4.06   0.000     1.888262    5.420601
    education |   1.683351   .7590775     2.22   0.027     .1944695    3.172232
income_norm01 |   .0583352   .4785026     0.12   0.903    -.8802166    .9968869
        _cons |  -7.810195   1.381165    -5.65   0.000    -10.51926    -5.10113
-------------------------------------------------------------------------------

. outreg2 using tabRepfear16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRepfear16.doc
dir : seeout

. 
. *********************

. ** Table B17 -- 2016 **

. *********************

. 
. svy: logit persuade c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,601.3724
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       9.33
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   .1361064   .4260928     0.32   0.749    -.6995737    .9717865
        1.web |  -.3032283    .287347    -1.06   0.291    -.8667914    .2603349
              |
web#c.hopeful |
           1  |   .6735055   .4901369     1.37   0.170     -.287782    1.634793
              |
  campaignint |   1.306806   .2634932     4.96   0.000     .7900261    1.823585
       polint |   .5814496   .3324745     1.75   0.080    -.0706204     1.23352
      polknow |   .7220862   .2595895     2.78   0.005     .2129628     1.23121
       gender |    .137729   .1398396     0.98   0.325    -.1365334    .4119914
        White |   .2892653   .1601029     1.81   0.071    -.0247386    .6032692
       latinx |   .0954457   .2359098     0.40   0.686    -.3672356    .5581269
         ageN |  -1.239783   .4196324    -2.95   0.003    -2.062793   -.4167734
    education |   .5244414   .4989881     1.05   0.293    -.4542057    1.503088
income_norm01 |   .0814906   .2622189     0.31   0.756    -.4327896    .5957708
        _cons |  -1.705009   .4335536    -3.93   0.000    -2.555322   -.8546964
-------------------------------------------------------------------------------

. outreg2 using tabDemhope16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemhope16.doc
dir : seeout

. svy: logit rally c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,827
Number of PSUs   = 1,827                          Population size = 1,600.4132
                                                  Design df       =      1,826
                                                  F(12, 1815)     =       3.64
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |    .269205   .6016896     0.45   0.655    -.9108671    1.449277
        1.web |   -.501993   .4709812    -1.07   0.287    -1.425711    .4217254
              |
web#c.hopeful |
           1  |   .6099212   .7227011     0.84   0.399    -.8074865    2.027329
              |
  campaignint |   1.348704   .5668378     2.38   0.017     .2369856    2.460423
       polint |   .8700613   .7054642     1.23   0.218    -.5135401    2.253663
      polknow |    .029332    .412461     0.07   0.943     -.779613    .8382769
       gender |   .0102101   .2130974     0.05   0.962    -.4077301    .4281503
        White |   .2526991   .2864712     0.88   0.378    -.3091465    .8145448
       latinx |  -.0104218   .4258905    -0.02   0.980    -.8457056    .8248619
         ageN |  -2.263524   .7266112    -3.12   0.002      -3.6886   -.8384472
    education |   .7743009   .8803316     0.88   0.379    -.9522617    2.500864
income_norm01 |  -.3958382   .4215935    -0.94   0.348    -1.222694    .4310178
        _cons |  -3.436698   .7188946    -4.78   0.000     -4.84664   -2.026756
-------------------------------------------------------------------------------

. outreg2 using tabDemhope16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemhope16.doc
dir : seeout

. svy: logit button c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,601.3724
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       5.72
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   .8293821   .6662628     1.24   0.213    -.4773346    2.136099
        1.web |   .1258311    .478871     0.26   0.793     -.813361    1.065023
              |
web#c.hopeful |
           1  |   .3466657   .7388875     0.47   0.639    -1.102487    1.795819
              |
  campaignint |   1.268425   .3676872     3.45   0.001     .5472939    1.989557
       polint |    .844476    .476834     1.77   0.077    -.0907211    1.779673
      polknow |   .5818281   .3198578     1.82   0.069    -.0454974    1.209154
       gender |   .0537348    .195809     0.27   0.784    -.3302982    .4377677
        White |  -.0105029   .2249023    -0.05   0.963    -.4515955    .4305897
       latinx |  -.4230612   .3334575    -1.27   0.205    -1.077059    .2309369
         ageN |  -1.709091   .6015349    -2.84   0.005    -2.888859   -.5293221
    education |  -.5191718   .6328139    -0.82   0.412    -1.760286    .7219429
income_norm01 |  -.0921781   .3429373    -0.27   0.788    -.7647684    .5804122
        _cons |  -3.127941   .6727051    -4.65   0.000    -4.447292   -1.808589
-------------------------------------------------------------------------------

. outreg2 using tabDemhope16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemhope16.doc
dir : seeout

. svy: logit volunteer c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,601.3724
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       5.03
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |    1.51993   1.099976     1.38   0.167     -.637413    3.677274
        1.web |   .4528109   .8951097     0.51   0.613    -1.302735    2.208357
              |
web#c.hopeful |
           1  |  -.3946697   1.269439    -0.31   0.756    -2.884374    2.095035
              |
  campaignint |   1.595283   .8279831     1.93   0.054    -.0286095    3.219176
       polint |   1.002481   1.013065     0.99   0.323    -.9844064    2.989368
      polknow |  -.0479273    .598585    -0.08   0.936     -1.22191    1.126056
       gender |  -.1966992   .3138286    -0.63   0.531    -.8121998    .4188014
        White |  -.5404239   .3219328    -1.68   0.093    -1.171819    .0909711
       latinx |  -.3678141   .5469851    -0.67   0.501    -1.440596    .7049678
         ageN |   -1.27079   .9093817    -1.40   0.162    -3.054327    .5127467
    education |   2.838672   1.130153     2.51   0.012     .6221452    5.055199
income_norm01 |  -.4149569   .5301813    -0.78   0.434    -1.454782    .6248682
        _cons |  -6.919331   1.363436    -5.07   0.000    -9.593388   -4.245275
-------------------------------------------------------------------------------

. outreg2 using tabDemhope16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemhope16.doc
dir : seeout

. svy: logit donatecand c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,826
Number of PSUs   = 1,826                          Population size = 1,599.5564
                                                  Design df       =      1,825
                                                  F(12, 1814)     =       6.65
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   .7816183   .6116851     1.28   0.201    -.4180581    1.981295
        1.web |    -.52328   .5473059    -0.96   0.339    -1.596692    .5501317
              |
web#c.hopeful |
           1  |   .2265315   .7880191     0.29   0.774    -1.318983    1.772046
              |
  campaignint |    .890789   .6317324     1.41   0.159    -.3482056    2.129784
       polint |   1.274628    .613812     2.08   0.038       .07078    2.478476
      polknow |   1.693979   .4263728     3.97   0.000     .8577487    2.530209
       gender |  -.1221725   .1761165    -0.69   0.488    -.4675836    .2232386
        White |  -.0314484   .2485604    -0.13   0.899    -.5189412    .4560444
       latinx |  -.3998073   .3792274    -1.05   0.292    -1.143573     .343958
         ageN |   .1904025   .6927301     0.27   0.783    -1.168225     1.54903
    education |    .931708   .8676572     1.07   0.283    -.7699974    2.633413
income_norm01 |   .4603792   .4073433     1.13   0.259    -.3385289    1.259287
        _cons |  -5.362785   .9256091    -5.79   0.000    -7.178149    -3.54742
-------------------------------------------------------------------------------

. outreg2 using tabDemhope16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDemhope16.doc
dir : seeout

. 
. *********************

. ** Table B18 -- 2016 **

. *********************

. 
. svy: logit persuade c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       8.64
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.459982   .4161464     3.51   0.000     .6437381    2.276226
        1.web |   .7188059   .2838777     2.53   0.011     .1619983    1.275614
              |
web#c.hopeful |
           1  |   -.521218   .4854416    -1.07   0.283     -1.47338    .4309441
              |
  campaignint |   1.074166   .2701491     3.98   0.000     .5442866    1.604046
       polint |   .4676157   .3500239     1.34   0.182    -.2189334    1.154165
      polknow |   .3780094   .2608411     1.45   0.147    -.1336134    .8896322
       gender |  -.1724494   .1411714    -1.22   0.222    -.4493479    .1044491
        White |    .131241   .2916823     0.45   0.653     -.440875    .7033569
       latinx |   .2532168   .4223173     0.60   0.549     -.575131    1.081565
         ageN |   .2319178   .4430294     0.52   0.601    -.6370555    1.100891
    education |  -.0388122   .5704184    -0.07   0.946    -1.157651    1.080026
income_norm01 |   .0107941   .2666961     0.04   0.968     -.512313    .5339012
        _cons |  -2.430441   .5231853    -4.65   0.000    -3.456634   -1.404247
-------------------------------------------------------------------------------

. outreg2 using tabRephope16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRephope16.doc
dir : seeout

. svy: logit rally c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       1.72
                                                  Prob > F        =     0.0572

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |    1.68418   .7792795     2.16   0.031     .1556736    3.212686
        1.web |   .6709636   .5999179     1.12   0.264    -.5057362    1.847663
              |
web#c.hopeful |
           1  |  -.7558446   .9049971    -0.84   0.404    -2.530938    1.019248
              |
  campaignint |    .454625   .5143869     0.88   0.377    -.5543114    1.463561
       polint |   .3925192   .7107622     0.55   0.581    -1.001595    1.786633
      polknow |   -.141039   .5495019    -0.26   0.797    -1.218851    .9367732
       gender |   .0857764   .2703107     0.32   0.751    -.4444204    .6159732
        White |  -.4844173   .5063124    -0.96   0.339    -1.477516    .5086816
       latinx |  -.4723906   .8373705    -0.56   0.573    -2.114838    1.170057
         ageN |  -1.750646   .9066714    -1.93   0.054    -3.529023    .0277305
    education |   1.083777   .9855829     1.10   0.272    -.8493793    3.016934
income_norm01 |   .9480187   .6519383     1.45   0.146    -.3307159    2.226753
        _cons |  -4.561626   1.145343    -3.98   0.000    -6.808142   -2.315111
-------------------------------------------------------------------------------

. outreg2 using tabRephope16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRephope16.doc
dir : seeout

. svy: logit button c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       4.20
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.504206   .7263726     2.07   0.039      .079473    2.928938
        1.web |    .499436   .5753836     0.87   0.386    -.6291415    1.628013
              |
web#c.hopeful |
           1  |   .1113768   .8808998     0.13   0.899    -1.616451    1.839204
              |
  campaignint |   .0881319   .4367513     0.20   0.840    -.7685274    .9447912
       polint |   1.152851   .5655995     2.04   0.042      .043464    2.262237
      polknow |   .2256328   .3660683     0.62   0.538    -.4923864     .943652
       gender |  -.1227527   .2108229    -0.58   0.560    -.5362681    .2907628
        White |    .458987   .4300857     1.07   0.286    -.3845981    1.302572
       latinx |   .7691046   .5668711     1.36   0.175    -.3427761    1.880985
         ageN |  -1.789211   .6697624    -2.67   0.008    -3.102906   -.4755152
    education |  -.8897199   .7199739    -1.24   0.217    -2.301902    .5224621
income_norm01 |  -1.064381   .3793637    -2.81   0.005    -1.808478   -.3202838
        _cons |  -2.600078   .7976534    -3.26   0.001    -4.164624   -1.035533
-------------------------------------------------------------------------------

. outreg2 using tabRephope16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRephope16.doc
dir : seeout

. svy: logit volunteer c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,333.9118
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       1.44
                                                  Prob > F        =     0.1408

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   1.095222   .7778162     1.41   0.159    -.4304138    2.620858
        1.web |   .8468814   .8462851     1.00   0.317    -.8130516    2.506814
              |
web#c.hopeful |
           1  |  -1.043466   1.231863    -0.85   0.397    -3.459684    1.372753
              |
  campaignint |   .6863256   .7081811     0.97   0.333    -.7027256    2.075377
       polint |  -.1726199   1.024477    -0.17   0.866    -2.182064    1.836825
      polknow |   .6355976   .6456333     0.98   0.325      -.63077    1.901965
       gender |  -.2270977   .3911746    -0.58   0.562    -.9943613    .5401658
        White |   .3439404   .7512905     0.46   0.647    -1.129667    1.817548
       latinx |   .1093747   1.165651     0.09   0.925    -2.176973    2.395723
         ageN |  -1.794645    1.33232    -1.35   0.178    -4.407904    .8186135
    education |  -.1927622   1.509517    -0.13   0.898    -3.153581    2.768056
income_norm01 |  -.1159552    .961492    -0.12   0.904    -2.001859    1.769949
        _cons |  -4.332179   1.836635    -2.36   0.018     -7.93462   -.7297392
-------------------------------------------------------------------------------

. outreg2 using tabRephope16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRephope16.doc
dir : seeout

. svy: logit donatecand c.hopeful##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,614
Number of PSUs   = 1,614                          Population size = 1,333.5352
                                                  Design df       =      1,613
                                                  F(12, 1602)     =       4.90
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
      hopeful |   3.308953   .7320798     4.52   0.000     1.873025     4.74488
        1.web |   1.088338   .6658827     1.63   0.102    -.2177486    2.394424
              |
web#c.hopeful |
           1  |  -1.557768   .9120505    -1.71   0.088    -3.346697    .2311604
              |
  campaignint |   1.525767   .6215417     2.45   0.014     .3066525    2.744881
       polint |  -.3112706   .5418025    -0.57   0.566    -1.373981    .7514401
      polknow |   1.107635   .3638865     3.04   0.002     .3938952    1.821375
       gender |   .0162663   .2189162     0.07   0.941    -.4131237    .4456564
        White |  -.7746093    .381251    -2.03   0.042    -1.522409     -.02681
       latinx |   .1918209   .5178079     0.37   0.711     -.823826    1.207468
         ageN |   3.693247   .9392484     3.93   0.000     1.850972    5.535523
    education |   2.154395   .7894108     2.73   0.006     .6060167    3.702774
income_norm01 |   .0597735   .4988752     0.12   0.905    -.9187381    1.038285
        _cons |  -8.828069   1.423346    -6.20   0.000    -11.61987   -6.036268
-------------------------------------------------------------------------------

. outreg2 using tabRephope16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabRephope16.doc
dir : seeout

. 
. *********************

. ** Table B19 -- 2016 **

. *********************

. 
. svy: logit persuade c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,829
Number of PSUs   = 1,829                          Population size = 1,602.0694
                                                  Design df       =      1,828
                                                  F(12, 1817)     =       9.81
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .3607271   .4068033     0.89   0.375    -.4371211    1.158575
        1.web |  -.2496421   .2546311    -0.98   0.327    -.7490404    .2497563
              |
  web#c.proud |
           1  |   .6038168   .4623257     1.31   0.192    -.3029254    1.510559
              |
  campaignint |   1.254036   .2656715     4.72   0.000     .7329847    1.775088
       polint |   .5863403   .3312848     1.77   0.077    -.0633961    1.236077
      polknow |   .6952423   .2539962     2.74   0.006      .197089    1.193396
       gender |     .12055   .1407967     0.86   0.392    -.1555893    .3966893
        White |   .3361357   .1593803     2.11   0.035      .023549    .6487223
       latinx |   .1101428   .2352763     0.47   0.640    -.3512957    .5715813
         ageN |  -1.272959   .4176813    -3.05   0.002    -2.092142   -.4537761
    education |   .5417784   .4969935     1.09   0.276    -.4329564    1.516513
income_norm01 |   .1116151   .2609443     0.43   0.669    -.4001652    .6233954
        _cons |  -1.799246   .4124475    -4.36   0.000    -2.608164   -.9903282
-------------------------------------------------------------------------------

. outreg2 using tabDempride16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDempride16.doc
dir : seeout

. svy: logit rally c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,828
Number of PSUs   = 1,828                          Population size = 1,601.1102
                                                  Design df       =      1,827
                                                  F(12, 1816)     =       3.38
                                                  Prob > F        =     0.0001

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |  -.0997543   .5760057    -0.17   0.863    -1.229453    1.029944
        1.web |  -.6589086   .4468541    -1.47   0.141    -1.535307    .2174899
              |
  web#c.proud |
           1  |   .9382044   .7010134     1.34   0.181    -.4366675    2.313076
              |
  campaignint |   1.380935   .5664664     2.44   0.015     .2699451    2.491925
       polint |   .8781565    .700083     1.25   0.210    -.4948906    2.251204
      polknow |   .0483596   .4183726     0.12   0.908    -.7721791    .8688984
       gender |   .0103827   .2127526     0.05   0.961    -.4068812    .4276467
        White |    .254961   .2848887     0.89   0.371    -.3037807    .8137026
       latinx |  -.0002529    .425742    -0.00   1.000     -.835245    .8347391
         ageN |  -2.217516   .7315318    -3.03   0.002    -3.652243   -.7827898
    education |   .7327604   .8918659     0.82   0.411    -1.016423    2.481944
income_norm01 |  -.3754031   .4209629    -0.89   0.373    -1.201022    .4502159
        _cons |  -3.285185    .726677    -4.52   0.000     -4.71039   -1.859981
-------------------------------------------------------------------------------

. outreg2 using tabDempride16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDempride16.doc
dir : seeout

. svy: logit button c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,829
Number of PSUs   = 1,829                          Population size = 1,602.0694
                                                  Design df       =      1,828
                                                  F(12, 1817)     =       5.83
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .4768511    .587723     0.81   0.417     -.675828     1.62953
        1.web |  -.0976409   .4287863    -0.23   0.820    -.9386033    .7433215
              |
  web#c.proud |
           1  |   .7719499   .6669342     1.16   0.247    -.5360832    2.079983
              |
  campaignint |   1.282256   .3757057     3.41   0.001     .5453985    2.019113
       polint |   .8418282   .4762629     1.77   0.077    -.0922484    1.775905
      polknow |   .5599022   .3255165     1.72   0.086    -.0785211    1.198325
       gender |   .0359035   .1957203     0.18   0.854    -.3479554    .4197625
        White |   .0262372   .2281241     0.12   0.908    -.4211741    .4736485
       latinx |  -.4005062   .3351121    -1.20   0.232    -1.057749    .2567366
         ageN |  -1.673786   .6042201    -2.77   0.006     -2.85882   -.4887519
    education |  -.5545857   .6382383    -0.87   0.385    -1.806339    .6971671
income_norm01 |  -.0562145   .3431002    -0.16   0.870    -.7291241    .6166952
        _cons |  -2.945743    .650689    -4.53   0.000    -4.221915   -1.669571
-------------------------------------------------------------------------------

. outreg2 using tabDempride16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDempride16.doc
dir : seeout

. svy: logit volunteer c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,829
Number of PSUs   = 1,829                          Population size = 1,602.0694
                                                  Design df       =      1,828
                                                  F(12, 1817)     =       4.78
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |    .678995   .8245928     0.82   0.410    -.9382479    2.296238
        1.web |   .1033241   .7048444     0.15   0.883    -1.279061    1.485709
              |
  web#c.proud |
           1  |   .1911239   1.022366     0.19   0.852    -1.814003    2.196251
              |
  campaignint |   1.667683   .8349129     2.00   0.046     .0301999    3.305167
       polint |   1.019176    .994903     1.02   0.306    -.9320898    2.970442
      polknow |   -.023572   .6020045    -0.04   0.969    -1.204261    1.157117
       gender |  -.1867359   .3077031    -0.61   0.544    -.7902224    .4167507
        White |  -.5388168   .3301431    -1.63   0.103    -1.186314    .1086805
       latinx |  -.3343066   .5519556    -0.61   0.545    -1.416837    .7482233
         ageN |  -1.179909   .8952638    -1.32   0.188    -2.935756    .5759388
    education |   2.728966   1.148486     2.38   0.018     .4764831    4.981449
income_norm01 |  -.3765916   .5268293    -0.71   0.475    -1.409842     .656659
        _cons |  -6.472019   1.263819    -5.12   0.000    -8.950699   -3.993338
-------------------------------------------------------------------------------

. outreg2 using tabDempride16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDempride16.doc
dir : seeout

. svy: logit donatecand c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,827
Number of PSUs   = 1,827                          Population size = 1,600.2534
                                                  Design df       =      1,826
                                                  F(12, 1815)     =       7.06
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.369704   .5830731     2.35   0.019     .2261435    2.513264
        1.web |  -.2335169   .4778319    -0.49   0.625    -1.170671    .7036376
              |
  web#c.proud |
           1  |  -.1887099   .7149775    -0.26   0.792     -1.59097     1.21355
              |
  campaignint |   .8593972   .6345819     1.35   0.176    -.3851855     2.10398
       polint |    1.21567   .6004422     2.02   0.043      .038044    2.393295
      polknow |   1.613927   .4108253     3.93   0.000     .8081906    2.419664
       gender |   -.174028   .1788531    -0.97   0.331    -.5248061    .1767502
        White |   .0567449   .2502522     0.23   0.821    -.4340658    .5475556
       latinx |  -.3375818   .3783099    -0.89   0.372    -1.079547    .4043838
         ageN |   .1800994   .6886952     0.26   0.794    -1.170614    1.530813
    education |   .9746166    .831023     1.17   0.241    -.6552388    2.604472
income_norm01 |   .4892371   .3997362     1.22   0.221     -.294751    1.273225
        _cons |  -5.686476   .8528982    -6.67   0.000    -7.359234   -4.013717
-------------------------------------------------------------------------------

. outreg2 using tabDempride16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabDempride16.doc
dir : seeout

. 
. *********************

. ** Table B20 -- 2016 **

. *********************

. 
. svy: logit persuade c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      8.50
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
     persuade | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.435456   .4104303     3.50   0.000     .6304245    2.240488
        1.web |   .6689749   .2473547     2.70   0.007     .1838051    1.154145
              |
  web#c.proud |
           1  |  -.5767897   .4820625    -1.20   0.232    -1.522323     .368744
              |
  campaignint |   1.089415   .2647793     4.11   0.000     .5700677    1.608762
       polint |   .5204814   .3543072     1.47   0.142    -.1744687    1.215432
      polknow |   .3545316   .2616973     1.35   0.176    -.1587704    .8678336
       gender |  -.1807914   .1407664    -1.28   0.199    -.4568954    .0953126
        White |   .1344072    .291153     0.46   0.644    -.4366701    .7054844
       latinx |   .1935115   .4212028     0.46   0.646    -.6326501    1.019673
         ageN |   .2183228   .4442107     0.49   0.623    -.6529672    1.089613
    education |  -.0549582    .575541    -0.10   0.924    -1.183844    1.073928
income_norm01 |  -.0011341   .2683796    -0.00   0.997     -.527543    .5252748
        _cons |  -2.274077   .5393439    -4.22   0.000    -3.331964   -1.216189
-------------------------------------------------------------------------------

. outreg2 using tabReppride16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabReppride16.doc
dir : seeout

. svy: logit rally c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      2.25
                                                   Prob > F        =    0.0082

-------------------------------------------------------------------------------
              |             Linearized
        rally | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   1.118906   .6797623     1.65   0.100    -.2144032    2.452215
        1.web |   .2178037     .52042     0.42   0.676    -.8029659    1.238573
              |
  web#c.proud |
           1  |  -.0364718   .7999974    -0.05   0.964    -1.605614     1.53267
              |
  campaignint |   .4415638   .5077533     0.87   0.385    -.5543607    1.437488
       polint |   .4049589   .7105658     0.57   0.569    -.9887689    1.798687
      polknow |  -.1420676   .5471596    -0.26   0.795    -1.215285    .9311498
       gender |   .0608763   .2760471     0.22   0.825    -.4805717    .6023244
        White |  -.4574348   .5086859    -0.90   0.369    -1.455189     .540319
       latinx |  -.4664814   .8414519    -0.55   0.579    -2.116934    1.183971
         ageN |  -1.703576   .8813739    -1.93   0.053    -3.432332    .0251812
    education |   1.110309   .9717493     1.14   0.253    -.7957129    3.016331
income_norm01 |   .9445974   .6449425     1.46   0.143    -.3204148     2.20961
        _cons |    -4.1658   1.230556    -3.39   0.001    -6.579454   -1.752145
-------------------------------------------------------------------------------

. outreg2 using tabReppride16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabReppride16.doc
dir : seeout

. svy: logit button c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      5.34
                                                   Prob > F        =    0.0000

-------------------------------------------------------------------------------
              |             Linearized
       button | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   2.043422   .6446126     3.17   0.002     .7790574    3.307787
        1.web |      .5949   .4914719     1.21   0.226    -.3690897     1.55889
              |
  web#c.proud |
           1  |  -.2202534   .8035801    -0.27   0.784    -1.796423    1.355916
              |
  campaignint |   .0586653   .4473535     0.13   0.896    -.8187891    .9361197
       polint |   1.112354   .5785055     1.92   0.055     -.022346    2.247055
      polknow |   .2635325    .372169     0.71   0.479    -.4664523    .9935173
       gender |  -.1379658   .2129198    -0.65   0.517     -.555594    .2796623
        White |   .4628346   .4461563     1.04   0.300    -.4122715    1.337941
       latinx |   .7421812   .5850172     1.27   0.205    -.4052914    1.889654
         ageN |  -1.724509   .6829561    -2.53   0.012    -3.064082   -.3849355
    education |    -.68319   .7169918    -0.95   0.341    -2.089522    .7231421
income_norm01 |  -1.049345    .380411    -2.76   0.006    -1.795496   -.3031938
        _cons |  -2.833774   .7451216    -3.80   0.000     -4.29528   -1.372267
-------------------------------------------------------------------------------

. outreg2 using tabReppride16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabReppride16.doc
dir : seeout

. svy: logit volunteer c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                           Number of obs   =     1,616
Number of PSUs   = 1,616                           Population size = 1,334.398
                                                   Design df       =     1,615
                                                   F(12, 1604)     =      1.24
                                                   Prob > F        =    0.2523

-------------------------------------------------------------------------------
              |             Linearized
    volunteer | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |   .6342418   .7151073     0.89   0.375    -.7683938    2.036878
        1.web |   .6196687   .7306552     0.85   0.397    -.8134633    2.052801
              |
  web#c.proud |
           1  |  -.7632938   1.186978    -0.64   0.520    -3.091473    1.564885
              |
  campaignint |   .7131684   .7018031     1.02   0.310    -.6633721    2.089709
       polint |  -.1006194   1.023636    -0.10   0.922    -2.108414    1.907175
      polknow |    .604514   .6559796     0.92   0.357    -.6821468    1.891175
       gender |  -.2285608   .3974637    -0.58   0.565     -1.00816     .551038
        White |   .3594577   .7471342     0.48   0.630    -1.105997    1.824912
       latinx |   .0965286   1.163896     0.08   0.934    -2.186377    2.379434
         ageN |  -1.798535   1.359459    -1.32   0.186    -4.465024    .8679537
    education |  -.2865663   1.513736    -0.19   0.850     -3.25566    2.682528
income_norm01 |  -.1361659   .9608636    -0.14   0.887    -2.020837    1.748505
        _cons |  -4.006839   1.738423    -2.30   0.021    -7.416642   -.5970368
-------------------------------------------------------------------------------

. outreg2 using tabReppride16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabReppride16.doc
dir : seeout

. svy: logit donatecand c.proud##i.web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1 
(running logit on estimation sample)

Survey: Logistic regression

Number of strata =     1                          Number of obs   =      1,615
Number of PSUs   = 1,615                          Population size = 1,334.0214
                                                  Design df       =      1,614
                                                  F(12, 1603)     =       4.17
                                                  Prob > F        =     0.0000

-------------------------------------------------------------------------------
              |             Linearized
   donatecand | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
        proud |    2.35351   .6478372     3.63   0.000      1.08282    3.624201
        1.web |   .3990849   .5043162     0.79   0.429    -.5900985    1.388268
              |
  web#c.proud |
           1  |  -.6424613   .7966308    -0.81   0.420    -2.205001    .9200781
              |
  campaignint |   1.596381   .6250526     2.55   0.011     .3703811    2.822381
       polint |  -.2231292   .5475218    -0.41   0.684    -1.297057    .8507992
      polknow |   1.092613   .3638804     3.00   0.003      .378885     1.80634
       gender |   .0391974   .2186992     0.18   0.858    -.3897667    .4681616
        White |  -.6805162   .3723503    -1.83   0.068    -1.410857    .0498246
       latinx |   .2017753   .5389388     0.37   0.708     -.855318    1.258869
         ageN |   3.660744   .9274979     3.95   0.000     1.841517     5.47997
    education |   2.162234   .7644354     2.83   0.005     .6628433    3.661624
income_norm01 |   .0305121   .4876194     0.06   0.950    -.9259216    .9869457
        _cons |  -8.147327   1.377482    -5.91   0.000    -10.84917   -5.445485
-------------------------------------------------------------------------------

. outreg2 using tabReppride16.doc, paren(se) bdec(2) sdec(2) alpha(.001, .01, .05) symbol(***,**,*) append
tabReppride16.doc
dir : seeout

. 
. *********************

. ** Figure C2 -- 2016 **

. *********************

. 
. svyset [pweight=V160101]

Sampling weights: V160101
             VCE: linearized
     Single unit: missing
        Strata 1: <one>
 Sampling unit 1: <observations>
           FPC 1: <zero>

. 
. svy: reg angrycandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,788
Number of PSUs   = 1,788                          Population size = 1,825.4169
                                                  Design df       =      1,787
                                                  F(10, 1778)     =       5.71
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0378

-------------------------------------------------------------------------------
              |             Linearized
 angrycandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0494662   .0118564     4.17   0.000     .0262124      .07272
  campaignint |   .0167039   .0216651     0.77   0.441    -.0257877    .0591956
       polint |    .052375   .0264525     1.98   0.048     .0004938    .1042561
      polknow |    -.02213   .0201939    -1.10   0.273     -.061736    .0174761
       gender |   .0265001   .0105238     2.52   0.012     .0058599    .0471404
        White |   .0026453   .0124351     0.21   0.832    -.0217436    .0270343
       latinx |   .0222648   .0173338     1.28   0.199    -.0117317    .0562614
         ageN |  -.0871944    .033006    -2.64   0.008    -.1519288   -.0224601
    education |   .0136606   .0374808     0.36   0.716    -.0598502    .0871713
income_norm01 |   .0364244   .0201315     1.81   0.071    -.0030595    .0759082
        _cons |   .3731641   .0309359    12.06   0.000     .3124898    .4338384
-------------------------------------------------------------------------------

. estimates store angryDavg

. svy: reg angrycandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,576
Number of PSUs   = 1,576                          Population size = 1,538.0834
                                                  Design df       =      1,575
                                                  F(10, 1566)     =       2.67
                                                  Prob > F        =     0.0031
                                                  R-squared       =     0.0237

-------------------------------------------------------------------------------
              |             Linearized
 angrycandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0258084   .0127978     2.02   0.044     .0007058    .0509109
  campaignint |  -.0110652   .0228184    -0.48   0.628    -.0558229    .0336925
       polint |   .0477451   .0291988     1.64   0.102    -.0095274    .1050177
      polknow |    .032877   .0202294     1.63   0.104    -.0068025    .0725565
       gender |   .0385566   .0111564     3.46   0.001     .0166736    .0604396
        White |  -.0007888   .0279348    -0.03   0.977    -.0555821    .0540046
       latinx |   .0218033   .0367714     0.59   0.553    -.0503228    .0939294
         ageN |  -.0791406   .0343782    -2.30   0.021    -.1465723   -.0117088
    education |  -.0164397   .0395609    -0.42   0.678    -.0940374    .0611579
income_norm01 |   .0098944   .0221093     0.45   0.655    -.0334722    .0532611
        _cons |   .4252115   .0445865     9.54   0.000     .3377563    .5126666
-------------------------------------------------------------------------------

. estimates store angryRavg

. coefplot (angryDavg) || (angryRavg), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN education income_norm01) xlabel(-.2 "-.2" -.1 "-.1"
>  0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Anger toward Presidential Candidates") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(angeravg16)

. 
. svy: reg afraidcandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,785
Number of PSUs   = 1,785                          Population size = 1,826.8326
                                                  Design df       =      1,784
                                                  F(10, 1775)     =       4.40
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0322

-------------------------------------------------------------------------------
              |             Linearized
afraidcandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0704361   .0126339     5.58   0.000     .0456572     .095215
  campaignint |   .0292281   .0214822     1.36   0.174    -.0129048    .0713611
       polint |   .0087252   .0279923     0.31   0.755     -.046176    .0636264
      polknow |  -.0416678   .0213131    -1.96   0.051     -.083469    .0001335
       gender |   .0075541   .0116452     0.65   0.517    -.0152857    .0303938
        White |   .0026437   .0133314     0.20   0.843     -.023503    .0287904
       latinx |   .0063234   .0198413     0.32   0.750    -.0325913    .0452381
         ageN |   .0125593    .033432     0.38   0.707    -.0530107    .0781294
    education |   .0695788   .0384808     1.81   0.071    -.0058934     .145051
income_norm01 |  -.0071064   .0218816    -0.32   0.745    -.0500227      .03581
        _cons |   .3091134   .0323763     9.55   0.000      .245614    .3726128
-------------------------------------------------------------------------------

. estimates store afraidDavg

. svy: reg afraidcandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,579
Number of PSUs   = 1,579                          Population size = 1,539.3775
                                                  Design df       =      1,578
                                                  F(10, 1569)     =       2.11
                                                  Prob > F        =     0.0210
                                                  R-squared       =     0.0203

-------------------------------------------------------------------------------
              |             Linearized
afraidcandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0442285   .0140686     3.14   0.002     .0166334    .0718235
  campaignint |  -.0355804   .0251304    -1.42   0.157    -.0848728     .013712
       polint |   .0565864   .0300911     1.88   0.060    -.0024363    .1156091
      polknow |   .0183993    .024601     0.75   0.455    -.0298547    .0666534
       gender |    .027366   .0128865     2.12   0.034     .0020895    .0526426
        White |  -.0203988   .0268395    -0.76   0.447    -.0730436    .0322459
       latinx |  -.0198859   .0418522    -0.48   0.635    -.1019778    .0622059
         ageN |  -.0452607   .0402575    -1.12   0.261    -.1242246    .0337032
    education |   -.057336   .0481463    -1.19   0.234    -.1517736    .0371015
income_norm01 |   .0134532   .0248557     0.54   0.588    -.0353005     .062207
        _cons |   .4324908    .048401     8.94   0.000     .3375538    .5274277
-------------------------------------------------------------------------------

. estimates store afraidRavg

. coefplot (afraidDavg) || (afraidRavg), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN education income_norm01) xlabel(-.2 "-.2" -.1 "-.
> 1" 0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Fear toward Presidential Candidates") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(fearavg16)

. 
. svy: reg hopecandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,789
Number of PSUs   = 1,789                          Population size = 1,829.2874
                                                  Design df       =      1,788
                                                  F(10, 1779)     =      11.30
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0787

-------------------------------------------------------------------------------
              |             Linearized
  hopecandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0019801   .0105567     0.19   0.851    -.0187247     .022685
  campaignint |   .1107696   .0185551     5.97   0.000     .0743777    .1471616
       polint |  -.0099352   .0239982    -0.41   0.679    -.0570027    .0371323
      polknow |  -.0276494   .0182052    -1.52   0.129     -.063355    .0080562
       gender |   .0098554   .0099901     0.99   0.324     -.009738    .0294488
        White |  -.0285159   .0122294    -2.33   0.020    -.0525013   -.0045304
       latinx |   -.002819   .0179177    -0.16   0.875    -.0379609    .0323228
         ageN |   .1053326   .0284742     3.70   0.000     .0494864    .1611789
    education |  -.0636001   .0326461    -1.95   0.052    -.1276287    .0004285
income_norm01 |  -.0322394   .0176691    -1.82   0.068    -.0668937    .0024148
        _cons |   .2709415   .0285806     9.48   0.000     .2148867    .3269964
-------------------------------------------------------------------------------

. estimates store hopefulDavg

. svy: reg hopecandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,578                          Population size = 1,538.8918
                                                  Design df       =      1,577
                                                  F(10, 1568)     =      12.28
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0944

-------------------------------------------------------------------------------
              |             Linearized
  hopecandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0293256   .0109385     2.68   0.007       .00787    .0507812
  campaignint |   .0889803   .0188398     4.72   0.000     .0520265     .125934
       polint |   .0496784   .0244636     2.03   0.042     .0016939    .0976629
      polknow |  -.0468439   .0158743    -2.95   0.003    -.0779808    -.015707
       gender |  -.0128933   .0096217    -1.34   0.180     -.031766    .0059795
        White |   .0078444   .0193512     0.41   0.685    -.0301123    .0458012
       latinx |   .0136855    .030129     0.45   0.650    -.0454116    .0727826
         ageN |   .0304746   .0303861     1.00   0.316    -.0291268     .090076
    education |  -.1482173   .0354214    -4.18   0.000    -.2176953   -.0787392
income_norm01 |  -.0452391   .0181728    -2.49   0.013    -.0808844   -.0095938
        _cons |   .3153737   .0359497     8.77   0.000     .2448594     .385888
-------------------------------------------------------------------------------

. estimates store hopefulRavg

. coefplot (hopefulDavg) || (hopefulRavg), scheme(plottig) legend(off) drop(?cons campaignint polint polknow ageN education income_norm01) xlabel(-.2 "-.2" -.1 "
> -.1" 0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Hope toward Presidential Candidates") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(hopeavg16)

. 
. svy: reg pridecandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==0
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,790
Number of PSUs   = 1,790                          Population size = 1,826.3782
                                                  Design df       =      1,789
                                                  F(10, 1780)     =      11.83
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0922

-------------------------------------------------------------------------------
              |             Linearized
 pridecandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |  -.0120943   .0107253    -1.13   0.260    -.0331298    .0089412
  campaignint |   .0988764   .0182266     5.42   0.000     .0631287    .1346241
       polint |   .0213401   .0239466     0.89   0.373    -.0256262    .0683064
      polknow |   .0111601   .0178548     0.63   0.532    -.0238582    .0461785
       gender |    .018385   .0096294     1.91   0.056     -.000501     .037271
        White |  -.0401337   .0116279    -3.45   0.001    -.0629394   -.0173279
       latinx |  -.0054931   .0169663    -0.32   0.746    -.0387689    .0277827
         ageN |   .0942444   .0290826     3.24   0.001     .0372049    .1512839
    education |  -.0366577   .0329069    -1.11   0.265    -.1011977    .0278822
income_norm01 |  -.0645814   .0192617    -3.35   0.001    -.1023592   -.0268035
        _cons |   .2317071   .0284216     8.15   0.000     .1759642    .2874501
-------------------------------------------------------------------------------

. estimates store proudDavg

. svy: reg pridecandavg web campaignint polint polknow gender White latinx ageN education income_norm01 if partyid_2cat==1
(running regress on estimation sample)

Survey: Linear regression

Number of strata =     1                          Number of obs   =      1,578
Number of PSUs   = 1,578                          Population size = 1,537.8061
                                                  Design df       =      1,577
                                                  F(10, 1568)     =      11.84
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.0952

-------------------------------------------------------------------------------
              |             Linearized
 pridecandavg | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
          web |   .0383196   .0106749     3.59   0.000      .017381    .0592581
  campaignint |   .0929649   .0191384     4.86   0.000     .0554254    .1305044
       polint |   .0385079   .0258226     1.49   0.136    -.0121423    .0891581
      polknow |   -.056458   .0175745    -3.21   0.001    -.0909299   -.0219862
       gender |  -.0046424   .0095759    -0.48   0.628    -.0234253    .0141405
        White |  -.0082489   .0216381    -0.38   0.703    -.0506914    .0341935
       latinx |  -.0247271   .0293771    -0.84   0.400    -.0823493    .0328952
         ageN |   .0185694   .0319339     0.58   0.561    -.0440679    .0812066
    education |  -.1545149   .0364868    -4.23   0.000    -.2260826   -.0829472
income_norm01 |  -.0390307    .018366    -2.13   0.034     -.075055   -.0030063
        _cons |   .2866445   .0383576     7.47   0.000     .2114073    .3618817
-------------------------------------------------------------------------------

. estimates store proudRavg

. coefplot (proudDavg) || (proudRavg), scheme(plottig) legend(off) drop(?cons ageN campaignint polint polknow education income_norm01) xlabel(-.2 "-.2" -.1 "-.1"
>  0 "0" .1 ".1" .2 ".2") xline(0) xtitle("Pride toward Presidential Candidates") mlabel format(%9.3f) mlabsize(medsmall) mlabposition(12) name(prideavg16)

. 
. graph combine angeravg16 fearavg16 hopeavg16 prideavg16, scheme(plottig) xcommon ycommon col(2)

. log close
      name:  <unnamed>
       log:  /Users/marziaoceno/Dropbox/How Social Desirability Bias Impacts the Expression of Emotions.log
  log type:  text
 closed on:  10 May 2025, 14:50:34
-----------------------------------------------------------------------------------------------------------------------------------------------------------------
